01_training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e85ad3db",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T17:52:42.354503Z",
     "iopub.status.busy": "2022-10-11T17:52:42.353695Z",
     "iopub.status.idle": "2022-10-11T17:52:47.681952Z",
     "shell.execute_reply": "2022-10-11T17:52:47.680987Z",
     "shell.execute_reply.started": "2022-10-11T17:52:42.354415Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(#2) [Path('../input/nike-adidas-and-converse-imaged/test'),Path('../input/nike-adidas-and-converse-imaged/train')]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os\n",
    "import gc\n",
    "from typing import NamedTuple, Optional\n",
    "from fastcore.all import *\n",
    "\n",
    "iskaggle = os.environ.get('KAGGLE_KERNEL_RUN_TYPE', '')\n",
    "if iskaggle:\n",
    "    import sys\n",
    "    sys.path.append(\"../input/timmmaster\")\n",
    "    path = Path(\"../input/nike-adidas-and-converse-imaged\")\n",
    "else:\n",
    "    path = Path(os.environ[\"PROJECTS\"])/\"shoes-classifier\"\n",
    "\n",
    "import timm\n",
    "from fastai.vision.all import *\n",
    "import torch\n",
    "\n",
    "(path).ls()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "93a6cf63",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T17:52:52.451914Z",
     "iopub.status.busy": "2022-10-11T17:52:52.451339Z",
     "iopub.status.idle": "2022-10-11T17:52:52.463378Z",
     "shell.execute_reply": "2022-10-11T17:52:52.461235Z",
     "shell.execute_reply.started": "2022-10-11T17:52:52.451879Z"
    }
   },
   "outputs": [],
   "source": [
    "def train(n_epochs, arch, item=None, batch=None, accum=None, bs=64, cbs=None, **fine_tune):\n",
    "    bs_accum = bs//accum if accum else bs\n",
    "    cbs = cbs if cbs is not None else []\n",
    "    if accum:\n",
    "        cbs.append(GradientAccumulation(bs))\n",
    "        \n",
    "    dls = ImageDataLoaders.from_folder(\n",
    "        path=path, train=\"train\", valid=\"test\", item_tfms=item, bs=bs_accum\n",
    "    )\n",
    "\n",
    "    learn = vision_learner(dls, arch, metrics=error_rate, cbs=cbs).to_fp16()\n",
    "    if iskaggle:\n",
    "        learn.path = Path(\"/kaggle/working\")\n",
    "    learn.fine_tune(n_epochs, 1e-2, **fine_tune)\n",
    "    return dls, learn\n",
    "\n",
    "def tta(learn):\n",
    "    preds, targs = learn.get_preds(dl=learn.dls.valid)\n",
    "    preds_tta, _ = learn.tta(dl=dls.valid)\n",
    "    return preds, preds_tta, targs\n",
    "\n",
    "def cleanup(tmp):\n",
    "    del tmp\n",
    "    gc.collect()\n",
    "    torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "33ce9fe2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T16:13:39.129760Z",
     "iopub.status.busy": "2022-10-11T16:13:39.129305Z",
     "iopub.status.idle": "2022-10-11T16:13:39.143191Z",
     "shell.execute_reply": "2022-10-11T16:13:39.142233Z",
     "shell.execute_reply.started": "2022-10-11T16:13:39.129720Z"
    }
   },
   "outputs": [],
   "source": [
    "frozen_epochs = (1, 5)\n",
    "batch_tfms = (\n",
    "    None,\n",
    "    [Rotate(max_deg=15)],\n",
    "    [Zoom(min_zoom=.8,max_zoom=1.2)],\n",
    "    [Rotate(max_deg=15), Zoom(min_zoom=.8,max_zoom=1.2)],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "85fc0816",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T15:00:11.008184Z",
     "iopub.status.busy": "2022-10-11T15:00:11.007741Z",
     "iopub.status.idle": "2022-10-11T15:22:36.604145Z",
     "shell.execute_reply": "2022-10-11T15:22:36.594179Z",
     "shell.execute_reply.started": "2022-10-11T15:00:11.008146Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "convnext_small_in22k 1 None\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://dl.fbaipublicfiles.com/convnext/convnext_small_22k_224.pth\" to /root/.cache/torch/hub/checkpoints/convnext_small_22k_224.pth\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.162607</td>\n",
       "      <td>1.896914</td>\n",
       "      <td>0.210526</td>\n",
       "      <td>00:15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.376832</td>\n",
       "      <td>0.470157</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.272804</td>\n",
       "      <td>0.440677</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.199492</td>\n",
       "      <td>0.301102</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.147125</td>\n",
       "      <td>0.240365</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.108879</td>\n",
       "      <td>0.388843</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.084527</td>\n",
       "      <td>0.268417</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.065818</td>\n",
       "      <td>0.339374</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.053402</td>\n",
       "      <td>0.333152</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.041952</td>\n",
       "      <td>0.322444</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.033537</td>\n",
       "      <td>0.323039</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 1 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.187275</td>\n",
       "      <td>0.690203</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.330733</td>\n",
       "      <td>0.445208</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.229800</td>\n",
       "      <td>0.263589</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.161758</td>\n",
       "      <td>0.213575</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.133437</td>\n",
       "      <td>0.371978</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.112579</td>\n",
       "      <td>0.420281</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.085685</td>\n",
       "      <td>0.342187</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.066611</td>\n",
       "      <td>0.306341</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.054101</td>\n",
       "      <td>0.268373</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.042106</td>\n",
       "      <td>0.261380</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.033732</td>\n",
       "      <td>0.260491</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 1 [Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.070994</td>\n",
       "      <td>0.719446</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.293765</td>\n",
       "      <td>0.429777</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.210733</td>\n",
       "      <td>0.353427</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.174465</td>\n",
       "      <td>0.285731</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.151558</td>\n",
       "      <td>0.273748</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.120382</td>\n",
       "      <td>0.281698</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.090989</td>\n",
       "      <td>0.300815</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.071860</td>\n",
       "      <td>0.315601</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.055865</td>\n",
       "      <td>0.333483</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.045773</td>\n",
       "      <td>0.336450</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.037035</td>\n",
       "      <td>0.338856</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 1 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: , Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.971787</td>\n",
       "      <td>1.077455</td>\n",
       "      <td>0.192982</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.238332</td>\n",
       "      <td>0.599896</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.185282</td>\n",
       "      <td>0.307706</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.150929</td>\n",
       "      <td>0.470835</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.142629</td>\n",
       "      <td>0.512461</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.158322</td>\n",
       "      <td>0.333709</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.129180</td>\n",
       "      <td>0.335106</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.104221</td>\n",
       "      <td>0.314975</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.081635</td>\n",
       "      <td>0.317827</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.064339</td>\n",
       "      <td>0.308994</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.051985</td>\n",
       "      <td>0.305032</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 5 None\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.344176</td>\n",
       "      <td>0.536693</td>\n",
       "      <td>0.175439</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.823378</td>\n",
       "      <td>0.268831</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.545024</td>\n",
       "      <td>0.289403</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.391121</td>\n",
       "      <td>0.393716</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.293803</td>\n",
       "      <td>0.343998</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.072101</td>\n",
       "      <td>0.298651</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.055179</td>\n",
       "      <td>0.229439</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.049456</td>\n",
       "      <td>0.370294</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.057559</td>\n",
       "      <td>0.479746</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.060078</td>\n",
       "      <td>0.447526</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.054612</td>\n",
       "      <td>0.460534</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.047446</td>\n",
       "      <td>0.476651</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.041694</td>\n",
       "      <td>0.478762</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.034258</td>\n",
       "      <td>0.473324</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.028318</td>\n",
       "      <td>0.466063</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 5 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.236385</td>\n",
       "      <td>1.039244</td>\n",
       "      <td>0.298246</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.824014</td>\n",
       "      <td>0.346944</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.560902</td>\n",
       "      <td>0.300532</td>\n",
       "      <td>0.061404</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.393739</td>\n",
       "      <td>0.335395</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.293676</td>\n",
       "      <td>0.346024</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.060262</td>\n",
       "      <td>0.236682</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.046843</td>\n",
       "      <td>0.289444</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.041744</td>\n",
       "      <td>0.333779</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.065109</td>\n",
       "      <td>0.362681</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.102969</td>\n",
       "      <td>0.263838</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.091236</td>\n",
       "      <td>0.171210</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.071267</td>\n",
       "      <td>0.164869</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.055929</td>\n",
       "      <td>0.153378</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.047094</td>\n",
       "      <td>0.157392</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.042229</td>\n",
       "      <td>0.156393</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 5 [Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.465643</td>\n",
       "      <td>1.027897</td>\n",
       "      <td>0.298246</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.956306</td>\n",
       "      <td>0.338434</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.632121</td>\n",
       "      <td>0.333956</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.443595</td>\n",
       "      <td>0.392369</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.337582</td>\n",
       "      <td>0.366310</td>\n",
       "      <td>0.122807</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.050482</td>\n",
       "      <td>0.268438</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.043175</td>\n",
       "      <td>0.280639</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.056215</td>\n",
       "      <td>0.266771</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.084606</td>\n",
       "      <td>0.529432</td>\n",
       "      <td>0.140351</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.099796</td>\n",
       "      <td>0.286614</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.084460</td>\n",
       "      <td>0.312953</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.067505</td>\n",
       "      <td>0.329499</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.053240</td>\n",
       "      <td>0.318879</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.042613</td>\n",
       "      <td>0.315730</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.035230</td>\n",
       "      <td>0.312978</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_small_in22k 5 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: , Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.295056</td>\n",
       "      <td>1.097856</td>\n",
       "      <td>0.350877</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.790534</td>\n",
       "      <td>0.358222</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.530578</td>\n",
       "      <td>0.397614</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.388907</td>\n",
       "      <td>0.359870</td>\n",
       "      <td>0.122807</td>\n",
       "      <td>00:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.296464</td>\n",
       "      <td>0.300830</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:09</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.062299</td>\n",
       "      <td>0.320162</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.052091</td>\n",
       "      <td>0.303831</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.057278</td>\n",
       "      <td>0.243529</td>\n",
       "      <td>0.061404</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.049608</td>\n",
       "      <td>0.334408</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.061070</td>\n",
       "      <td>0.410221</td>\n",
       "      <td>0.061404</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.071758</td>\n",
       "      <td>0.409641</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.068353</td>\n",
       "      <td>0.479451</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.055357</td>\n",
       "      <td>0.406124</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.044344</td>\n",
       "      <td>0.373390</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.036511</td>\n",
       "      <td>0.366302</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 1 None\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_224.pth\" to /root/.cache/torch/hub/checkpoints/convnext_large_22k_224.pth\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.925978</td>\n",
       "      <td>0.466902</td>\n",
       "      <td>0.131579</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      <progress value='0' class='' max='10' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      0.00% [0/10 00:00&lt;?]\n",
       "    </div>\n",
       "    \n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table><p>\n",
       "\n",
       "    <div>\n",
       "      <progress value='3' class='' max='22' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      13.64% [3/22 00:04&lt;00:27 0.2638]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "RuntimeError",
     "evalue": "CUDA out of memory. Tried to allocate 44.00 MiB (GPU 0; 15.90 GiB total capacity; 14.23 GiB already allocated; 63.75 MiB free; 15.00 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_17/3865628409.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      2\u001b[0m     \u001b[0macumm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m2\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;34m\"large\"\u001b[0m \u001b[0;32min\u001b[0m \u001b[0march\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0march\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 4\u001b[0;31m     \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0march\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0march\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mfreeze_epochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbatch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maccum\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0macumm\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      5\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"----------------------------------------------\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/tmp/ipykernel_17/1197844202.py\u001b[0m in \u001b[0;36mtrain\u001b[0;34m(n_epochs, arch, item, batch, accum, bs, **fine_tune)\u001b[0m\n\u001b[1;32m      8\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0miskaggle\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m         \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mPath\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"/kaggle/working\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 10\u001b[0;31m     \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfine_tune\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_epochs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1e-2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfine_tune\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     11\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mdls\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlearn\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/callback/schedule.py\u001b[0m in \u001b[0;36mfine_tune\u001b[0;34m(self, epochs, base_lr, freeze_epochs, lr_mult, pct_start, div, **kwargs)\u001b[0m\n\u001b[1;32m    166\u001b[0m     \u001b[0mbase_lr\u001b[0m \u001b[0;34m/=\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    167\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munfreeze\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 168\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_one_cycle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mepochs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mslice\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbase_lr\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mlr_mult\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbase_lr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpct_start\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mpct_start\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdiv\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdiv\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    169\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    170\u001b[0m \u001b[0;31m# %% ../nbs/14_callback.schedule.ipynb 68\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/callback/schedule.py\u001b[0m in \u001b[0;36mfit_one_cycle\u001b[0;34m(self, n_epoch, lr_max, div, div_final, pct_start, wd, moms, cbs, reset_opt, start_epoch)\u001b[0m\n\u001b[1;32m    117\u001b[0m     scheds = {'lr': combined_cos(pct_start, lr_max/div, lr_max, lr_max/div_final),\n\u001b[1;32m    118\u001b[0m               'mom': combined_cos(pct_start, *(self.moms if moms is None else moms))}\n\u001b[0;32m--> 119\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcbs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mParamScheduler\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mscheds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0mL\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcbs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreset_opt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mreset_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mwd\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstart_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mstart_epoch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    120\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    121\u001b[0m \u001b[0;31m# %% ../nbs/14_callback.schedule.ipynb 51\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, n_epoch, lr, wd, cbs, reset_opt, start_epoch)\u001b[0m\n\u001b[1;32m    254\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_hypers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlr\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mlr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    255\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_epoch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mn_epoch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 256\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_fit\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'fit'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelFitException\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_end_cleanup\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    257\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    258\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_end_cleanup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxb\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    191\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    192\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 193\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    194\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    195\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_fit\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    243\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    244\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 245\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'epoch'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelEpochException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    246\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    247\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcbs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreset_opt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstart_epoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    191\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    192\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 193\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    194\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    195\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_epoch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    237\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    238\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 239\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    240\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    241\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_epoch_train\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    229\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    230\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 231\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mall_batches\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'train'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelTrainException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    232\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    233\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mds_idx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdl\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    191\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    192\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 193\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    194\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    195\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mall_batches\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    197\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mall_batches\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    198\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 199\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mo\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mone_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    200\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    201\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_backward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_grad\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36mone_batch\u001b[0;34m(self, i, b)\u001b[0m\n\u001b[1;32m    225\u001b[0m         \u001b[0mb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_set_device\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    226\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_split\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 227\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_one_batch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'batch'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelBatchException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    228\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    229\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    191\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    192\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 193\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    194\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    195\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m  \u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/learner.py\u001b[0m in \u001b[0;36m_do_one_batch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    203\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    204\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 205\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    206\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'after_pred'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    207\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    139\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    140\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 141\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    142\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/fastai/vision/learner.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    175\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mcut\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mcut_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcut\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    176\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 177\u001b[0;31m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforward_features\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mneeds_pool\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    178\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    179\u001b[0m \u001b[0;31m# %% ../nbs/21_vision.learner.ipynb 37\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/kaggle/input/timmmaster/timm/models/convnext.py\u001b[0m in \u001b[0;36mforward_features\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    396\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward_features\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    397\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstem\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 398\u001b[0;31m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstages\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    399\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnorm_pre\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    400\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    139\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    140\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 141\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    142\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/kaggle/input/timmmaster/timm/models/convnext.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    248\u001b[0m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcheckpoint_seq\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mblocks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    249\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 250\u001b[0;31m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mblocks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    251\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    252\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/container.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    139\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    140\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mmodule\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 141\u001b[0;31m             \u001b[0minput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodule\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    142\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/kaggle/input/timmmaster/timm/models/convnext.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m    182\u001b[0m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpermute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    183\u001b[0m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnorm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 184\u001b[0;31m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmlp\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    185\u001b[0m             \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpermute\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    186\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgamma\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/kaggle/input/timmmaster/timm/models/layers/mlp.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, x)\u001b[0m\n\u001b[1;32m     26\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     27\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfc1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 28\u001b[0;31m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mact\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     29\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdrop1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     30\u001b[0m         \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfc2\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1108\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1109\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1110\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1111\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1112\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/kaggle/input/timmmaster/timm/models/layers/activations.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    144\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mforward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTensor\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 145\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mF\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgelu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mRuntimeError\u001b[0m: CUDA out of memory. Tried to allocate 44.00 MiB (GPU 0; 15.90 GiB total capacity; 14.23 GiB already allocated; 63.75 MiB free; 15.00 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF"
     ]
    }
   ],
   "source": [
    "arch = \"convnext_small_in22k\"\n",
    "for epochs, batch in itertools.product(frozen_epochs, batch_tfms):\n",
    "    print(arch, epochs, batch)\n",
    "    _ = train(10, arch=arch,freeze_epochs=epochs, batch=batch)\n",
    "    cleanup(_)\n",
    "    print(\"----------------------------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8b86c324",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T16:14:04.832008Z",
     "iopub.status.busy": "2022-10-11T16:14:04.831552Z",
     "iopub.status.idle": "2022-10-11T17:12:15.167163Z",
     "shell.execute_reply": "2022-10-11T17:12:15.166031Z",
     "shell.execute_reply.started": "2022-10-11T16:14:04.831967Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "convnext_large_in22k 1 None\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_224.pth\" to /root/.cache/torch/hub/checkpoints/convnext_large_22k_224.pth\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.960274</td>\n",
       "      <td>0.490255</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:31</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.190588</td>\n",
       "      <td>0.187724</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.169418</td>\n",
       "      <td>0.259254</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.144801</td>\n",
       "      <td>0.203997</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.125380</td>\n",
       "      <td>0.224654</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.089309</td>\n",
       "      <td>0.254844</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.070449</td>\n",
       "      <td>0.198235</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.046262</td>\n",
       "      <td>0.255114</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.030198</td>\n",
       "      <td>0.238775</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.024406</td>\n",
       "      <td>0.249626</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.013691</td>\n",
       "      <td>0.248247</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 1 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.805464</td>\n",
       "      <td>0.366086</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.234294</td>\n",
       "      <td>0.283707</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.159089</td>\n",
       "      <td>0.204957</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.170870</td>\n",
       "      <td>0.315931</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.168604</td>\n",
       "      <td>0.258700</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.131514</td>\n",
       "      <td>0.274895</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.089018</td>\n",
       "      <td>0.214492</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.056773</td>\n",
       "      <td>0.248511</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.037055</td>\n",
       "      <td>0.218487</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.025157</td>\n",
       "      <td>0.184959</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.016347</td>\n",
       "      <td>0.182398</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 1 [Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.826444</td>\n",
       "      <td>0.537221</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.285831</td>\n",
       "      <td>0.194487</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.196159</td>\n",
       "      <td>0.228085</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.180407</td>\n",
       "      <td>0.383271</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.143278</td>\n",
       "      <td>0.318576</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.114127</td>\n",
       "      <td>0.114509</td>\n",
       "      <td>0.026316</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.097290</td>\n",
       "      <td>0.387325</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.085590</td>\n",
       "      <td>0.152598</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.063079</td>\n",
       "      <td>0.160572</td>\n",
       "      <td>0.017544</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.047280</td>\n",
       "      <td>0.149088</td>\n",
       "      <td>0.026316</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.037184</td>\n",
       "      <td>0.147675</td>\n",
       "      <td>0.026316</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 1 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: , Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.896933</td>\n",
       "      <td>0.374916</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.329581</td>\n",
       "      <td>0.294623</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.222008</td>\n",
       "      <td>0.391124</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.150807</td>\n",
       "      <td>0.275523</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.169629</td>\n",
       "      <td>0.517870</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.133248</td>\n",
       "      <td>0.238507</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.105658</td>\n",
       "      <td>0.344152</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.075847</td>\n",
       "      <td>0.224426</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.044191</td>\n",
       "      <td>0.199824</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.029630</td>\n",
       "      <td>0.196163</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.023460</td>\n",
       "      <td>0.177238</td>\n",
       "      <td>0.026316</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 5 None\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.827139</td>\n",
       "      <td>0.307414</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.470444</td>\n",
       "      <td>0.224372</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.341474</td>\n",
       "      <td>0.278628</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.279001</td>\n",
       "      <td>0.271423</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.232021</td>\n",
       "      <td>0.353509</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.169276</td>\n",
       "      <td>0.189932</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.102864</td>\n",
       "      <td>0.197824</td>\n",
       "      <td>0.026316</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.108645</td>\n",
       "      <td>0.319419</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.126205</td>\n",
       "      <td>0.128844</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.105863</td>\n",
       "      <td>0.370102</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.084843</td>\n",
       "      <td>0.324707</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.078214</td>\n",
       "      <td>0.249515</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.052062</td>\n",
       "      <td>0.311507</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.042063</td>\n",
       "      <td>0.244660</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.031351</td>\n",
       "      <td>0.235307</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 5 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.787440</td>\n",
       "      <td>0.243620</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.402641</td>\n",
       "      <td>0.272142</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.265615</td>\n",
       "      <td>0.227051</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.240452</td>\n",
       "      <td>0.248925</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.209117</td>\n",
       "      <td>0.223743</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.113565</td>\n",
       "      <td>0.285723</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.143738</td>\n",
       "      <td>0.231032</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.141319</td>\n",
       "      <td>0.202138</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.178698</td>\n",
       "      <td>0.232541</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.127115</td>\n",
       "      <td>0.347957</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.088233</td>\n",
       "      <td>0.219896</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.054748</td>\n",
       "      <td>0.150788</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.040365</td>\n",
       "      <td>0.161689</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.028858</td>\n",
       "      <td>0.139744</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.023431</td>\n",
       "      <td>0.142241</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 5 [Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.834080</td>\n",
       "      <td>0.328552</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.414336</td>\n",
       "      <td>0.351220</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.274683</td>\n",
       "      <td>0.346493</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.244493</td>\n",
       "      <td>0.705495</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.263570</td>\n",
       "      <td>0.506179</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.128903</td>\n",
       "      <td>0.394496</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.132171</td>\n",
       "      <td>0.437772</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.206368</td>\n",
       "      <td>0.493983</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.187051</td>\n",
       "      <td>0.229428</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.130792</td>\n",
       "      <td>0.234329</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.118720</td>\n",
       "      <td>0.126806</td>\n",
       "      <td>0.017544</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.103220</td>\n",
       "      <td>0.139062</td>\n",
       "      <td>0.026316</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.067221</td>\n",
       "      <td>0.179188</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.050974</td>\n",
       "      <td>0.166233</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.039278</td>\n",
       "      <td>0.156992</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n",
      "convnext_large_in22k 5 [Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: , Zoom -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0}:\n",
      "encodes: (TensorImage,object) -> encodes\n",
      "(TensorMask,object) -> encodes\n",
      "(TensorBBox,object) -> encodes\n",
      "(TensorPoint,object) -> encodes\n",
      "decodes: ]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.778010</td>\n",
       "      <td>0.301766</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.446931</td>\n",
       "      <td>0.491511</td>\n",
       "      <td>0.114035</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.294870</td>\n",
       "      <td>0.341703</td>\n",
       "      <td>0.087719</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.281173</td>\n",
       "      <td>0.323632</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.287924</td>\n",
       "      <td>0.512204</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.133481</td>\n",
       "      <td>0.239891</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.084572</td>\n",
       "      <td>0.406082</td>\n",
       "      <td>0.070175</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.131919</td>\n",
       "      <td>0.441380</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.157601</td>\n",
       "      <td>0.466901</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.138538</td>\n",
       "      <td>0.337797</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.132053</td>\n",
       "      <td>0.223125</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.086902</td>\n",
       "      <td>0.248661</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.057313</td>\n",
       "      <td>0.257643</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.039333</td>\n",
       "      <td>0.253943</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.031119</td>\n",
       "      <td>0.273814</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "arch = \"convnext_large_in22k\"\n",
    "for epochs, batch in itertools.product(frozen_epochs, batch_tfms):\n",
    "    print(arch, epochs, batch)\n",
    "    _ = train(10, arch=arch,freeze_epochs=epochs, batch=batch, bs=66, accum=3)\n",
    "    cleanup(_)\n",
    "    print(\"----------------------------------------------\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2213898",
   "metadata": {},
   "source": [
    "## Train the final model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "939b8d67",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T17:54:02.051623Z",
     "iopub.status.busy": "2022-10-11T17:54:02.051110Z",
     "iopub.status.idle": "2022-10-11T18:02:21.321681Z",
     "shell.execute_reply": "2022-10-11T18:02:21.320328Z",
     "shell.execute_reply.started": "2022-10-11T17:54:02.051580Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading: \"https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_224.pth\" to /root/.cache/torch/hub/checkpoints/convnext_large_22k_224.pth\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.973437</td>\n",
       "      <td>0.328732</td>\n",
       "      <td>0.096491</td>\n",
       "      <td>00:30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.517086</td>\n",
       "      <td>0.227966</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.291923</td>\n",
       "      <td>0.164044</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.176703</td>\n",
       "      <td>0.223171</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.115359</td>\n",
       "      <td>0.251981</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Better model found at epoch 0 with error_rate value: 0.09649119526147842.\n",
      "Better model found at epoch 1 with error_rate value: 0.05263155326247215.\n",
      "Better model found at epoch 2 with error_rate value: 0.03508768603205681.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>error_rate</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.073164</td>\n",
       "      <td>0.242258</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.098828</td>\n",
       "      <td>0.274550</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.147813</td>\n",
       "      <td>0.310154</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.150457</td>\n",
       "      <td>0.155730</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.135424</td>\n",
       "      <td>0.202698</td>\n",
       "      <td>0.061403</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.092312</td>\n",
       "      <td>0.187794</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.061137</td>\n",
       "      <td>0.198197</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.039591</td>\n",
       "      <td>0.175930</td>\n",
       "      <td>0.035088</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.027868</td>\n",
       "      <td>0.175411</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.026567</td>\n",
       "      <td>0.184151</td>\n",
       "      <td>0.043860</td>\n",
       "      <td>00:32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Better model found at epoch 0 with error_rate value: 0.05263154208660126.\n",
      "Better model found at epoch 6 with error_rate value: 0.043859630823135376.\n",
      "Better model found at epoch 7 with error_rate value: 0.035087697207927704.\n"
     ]
    }
   ],
   "source": [
    "dls, learn = train(\n",
    "    10, freeze_epochs=5, arch=\"convnext_large_in22k\", accum=3, bs=66,\n",
    "    batch=[Zoom(min_zoom=.8,max_zoom=1.2)],\n",
    "    cbs=[SaveModelCallback(monitor=\"error_rate\")]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "828350ff",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T18:11:35.580577Z",
     "iopub.status.busy": "2022-10-11T18:11:35.580130Z",
     "iopub.status.idle": "2022-10-11T18:11:35.585884Z",
     "shell.execute_reply": "2022-10-11T18:11:35.584826Z",
     "shell.execute_reply.started": "2022-10-11T18:11:35.580538Z"
    }
   },
   "outputs": [],
   "source": [
    "from copy import deepcopy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "987ba4e8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T18:17:29.486772Z",
     "iopub.status.busy": "2022-10-11T18:17:29.486313Z",
     "iopub.status.idle": "2022-10-11T18:17:29.523094Z",
     "shell.execute_reply": "2022-10-11T18:17:29.522101Z",
     "shell.execute_reply.started": "2022-10-11T18:17:29.486734Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[0;31mSignature:\u001b[0m \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwith_opt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstrict\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mSource:\u001b[0m   \n",
       "\u001b[0;34m@\u001b[0m\u001b[0mpatch\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;34m@\u001b[0m\u001b[0mdelegates\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mload_model\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m\u001b[0;32mdef\u001b[0m \u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mLearner\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0;34m\"Load model and optimizer state (if `with_opt`) from `self.path/self.model_dir/file` using `device`\"\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0;32mif\u001b[0m \u001b[0mdevice\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdls\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'device'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mdevice\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_opt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mfile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjoin_path_file\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_dir\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mext\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'.pth'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfile\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdevice\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0mnested_attr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"accelerator.wait_for_everyone\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\n",
       "\u001b[0;34m\u001b[0m    \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
       "\u001b[0;31mFile:\u001b[0m      /opt/conda/lib/python3.7/site-packages/fastai/learner.py\n",
       "\u001b[0;31mType:\u001b[0m      method\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.load??"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a92f6e2a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T18:18:12.274678Z",
     "iopub.status.busy": "2022-10-11T18:18:12.274017Z",
     "iopub.status.idle": "2022-10-11T18:18:12.850305Z",
     "shell.execute_reply": "2022-10-11T18:18:12.849227Z",
     "shell.execute_reply.started": "2022-10-11T18:18:12.274612Z"
    }
   },
   "outputs": [],
   "source": [
    "best_learn = learn.load(\"/kaggle/working/models/model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "4703ffd0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T18:18:12.853506Z",
     "iopub.status.busy": "2022-10-11T18:18:12.852811Z",
     "iopub.status.idle": "2022-10-11T18:18:27.039366Z",
     "shell.execute_reply": "2022-10-11T18:18:27.038113Z",
     "shell.execute_reply.started": "2022-10-11T18:18:12.853468Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      <progress value='0' class='' max='10' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      \n",
       "    </div>\n",
       "    \n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(TensorBase(0.0351), TensorBase(0.0351))"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preds, preds_tta, targs = tta(best_learn)\n",
    "error_rate(preds, targs), error_rate(preds_tta, targs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "c93ef9cb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T18:18:27.047801Z",
     "iopub.status.busy": "2022-10-11T18:18:27.045140Z",
     "iopub.status.idle": "2022-10-11T18:18:33.114448Z",
     "shell.execute_reply": "2022-10-11T18:18:33.113306Z",
     "shell.execute_reply.started": "2022-10-11T18:18:27.047754Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<style>\n",
       "    /* Turns off some styling */\n",
       "    progress {\n",
       "        /* gets rid of default border in Firefox and Opera. */\n",
       "        border: none;\n",
       "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
       "        background-size: auto;\n",
       "    }\n",
       "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
       "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
       "    }\n",
       "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
       "        background: #F44336;\n",
       "    }\n",
       "</style>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABVMAAAC8CAYAAAB19z4VAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOy9d6AlyVne/XuruvukG+dO2p2ZzbvSroQCEkhCAQkwIIIRJnwEgwAbgwHjz8Bnm2SiMdjYYCwwweScBSIIMGgloZxWm3PeyXPnhhM6VdX3R1Wf0/fMvXfu7M7Mzmj7ke7OOd19uqur36queup531ecczRo0KBBgwYNGjRo0KBBgwYNGjRo0KBBg+2hnu0CNGjQoEGDBg0aNGjQoEGDBg0aNGjQoMHlgIZMbdCgQYMGDRo0aNCgQYMGDRo0aNCgQYMdoCFTGzRo0KBBgwYNGjRo0KBBgwYNGjRo0GAHaMjUBg0aNGjQoEGDBg0aNGjQoEGDBg0aNNgBGjK1QYMGDRo0aNCgQYMGDRo0aNCgQYMGDXaAhkxt0KBBgwYNGjRo0KBBgwYNGjRo0KBBgx2gIVMbNGjQoEGDBpc0RMSFvx8M319f2/b683D+XwvnevSZnusTHSLyaKirX3u2y3K54HzWmYh8Xc32rzmX4zaz8+Z5NmjQoEGDBg0anDsaMrVBgwYNGjRocFaIyK01csaJiBGRp0TkbSLyaRe5OGvAB8Lf2k5/NE3K1vBQONfHzlsJN7++FpGTIvKWqe0/XyvbERGJnsa5q+dz63kr8LmX4TYR+Yvw+dFnuzzTEJH/N9jtXhH5wSl7duHZvE9E/vmzXdbzgBNM2ki2zXEfC8c8VG1oCNYGDRo0aNCgQYPtcc6D9QYNGjRo0KDBcxo5noBpAS8EvgD4XBF5tXPug9MHB2LQOOfc+SqAc+6jwCvP4/l+BPiR83W+bfBaYAl4a7VBRDrAV9SO2Q98LvAXF6E85w1B/fhi4C1nOfTZxJuA9znnjotIffttQAnchLerV4rIXufc/9jqRCKSOOfyC1jWZwTn3F8Cf7mD4774IhSnQYMGDRo0aNDgEwqNMrVBgwYNGjRocC444px7pXPupXhyCvzi7FfV3YiDm/EjePJ1HkBEvkJE3i8ig/D3DyLy6vrJReS1QeGYisjHpveHYzZ18xeRl4nIW4PCMBORx0TkB6rja6f4gbq78xbuz1pEvlNE7grnWgvl/cwtyvF1IvIXIjIUkUdE5F9sUndvAlaAW2vbvjjUj2GijP36Te55X1CwPi4iuYgcF5G3hX0O+PRw6KfX3bu3cPe+pl7usO1qEflrEXlCREbh786g5pTp8mxxbxb48x0cW93TLhF5S7inItzT74rI9VP3/Zsicjg8h+Mi8u66elRE/p2I3B1sai08s1+dutYS8BpqRHYNX+yc+xTgBmAYtn1t+F29/r5cRD4sIjnweWH/a0Tkb0RkNZTvPhH5XhGJN79l+X4RORrK+jsiMl/b+V3B9pdDfZwQkT8RkZu2qMJbRORdoa08ICJfXDvXTsMBjFWolV0AV4fdb66d43Nqn2+u/f4bwraRiCxsdZ0GDRo0aNCgQYNPJDRkaoMGDRo0aNDgfONK4JfxROpxABH5TuB3gVcAR4BTwBuAd4jIq8Ix+4C/wiscHV79+lc7uaD4UAPvBb4ImAMeADrhGlVYgApPcXa3/l8AfhK4BXgi3MsbgL8VkTducvwvAi8ACuAa4BdF5PlTx3wR8JfOubK2rSJO/wb4n+HzF4rI7tq9LYXyfhNwCHgMGOBVwYR96+HzOjtz757GHrwiFuAefJ29APgp4Ft28Ps3EVSfO7mYiLSBdwLfireX+4FZvEr3/SJyMBz6c8A/xz/TO/Fk56cBrw/n+ULgfwA34+vlCXz9f83UJb8Q0GxOpu4UvwlcEa5TEfnvAD4bTyQ/ile3/ijwG5v8/kuA7wRWgS7wlcAv1fa/Hk/oHgXuBRbxZPvfh/qaxh8Ae/G2eQPwhyLySU//9sjwdlMpbk8ysaW/BR4M27+h9psvDf/+mXNu5Rlcu0GDBg0aNGjQ4LJBQ6Y2aNCgQYMGDc4FV4hXl36MCTFV4onSCjHwLc655+HJpwL4obDvvzjnbsATXn8bjv3hsO9bgBk8kfppzrlb8OTTTvCjQIInql7snHshsA/4d865jzrn6mEB/k9Q127q4hyUkRVh9LOhvNfiCVoVrjWNPwOuw7vyE457fe2cLwn3/NbatkPAZ4SvvwH8MZ4sjIGvrp37W5moBb/aOXejc+5a4GUA4d4+GvZ/NNzbK51zRza7vy3wIHCtc+6Qc+6T8c/tXWHfV2z9s7OqPrfCV+LDRAB8hXPuBcCn4BW6u4F/F/ZVqsx/7Zx7mXPuGvxz/V9T+//eOXdLOM88k3qt8CbgLufcg5yJPxWRD+LroBu2bUaG/hFwyDl3Iz4Mww/hVdlPANcFe/+J6p42ITYz4KZwXEWcf6mIXBc+/wdgMdzHJzEhtw8CZyi0gZ92zj0fT/gP8GTxd21y3I7gnDsSbKmym7+s2ZLDE9sAXyMiUVDVVkrtX3u6123QoEGDBg0aNLjc0JCpDRo0aNCgQYNzQYJXl74In+TmL4FPd87VlZ8jguIukDC3AL2w77uDK7HBK/pgEv+0Ip8ecM5VqtHf32G5XhH+/VPn3D3VtWvnORe8DKhc238nnGudSRzTl4iInvrNb4d7vbu2bV/t85uAFHh7bdub8WOxVbyyr8+EkKy7+lf39qhz7neqjSF27PlCAfx78aERCvzzeV3Yd+VZfvt0VJ+fEv7N8SQyzrk7gdvD9peHf98W/v01EXlIRP4Kr9A9HLb/TTjHZ4oP7/Be4GfC/QAgIl28rW1VvpeE8pTA+4Gv3SJe6v9yztlQVlO7h7c7506Hz79TO/7lbMStNeVuZdeCVwADXIVXaq+JiAX+rvbbzZ7B74eyPAm8J2x74SbHnS/8Kp7s34dXRf9TfH9wmI1lbdCgQYMGDRo0+IRGk4CqQYMGDRo0aHAueCyoA7fDiYp02gT34snDOs5bcqpnESsAzrmyFmK0Hmv0TXj1ZL+27evCvzPA0fC7Vtj2YhF5iXPutmdYrnrdVgTw/CbH/TTwL8PnB4Bl4Hq8SnSaOJ7Gm9ha9flM8b14ovBz8ETha4A3Al8GvMQ5d6eIvAD4KuCl+BAR/xr4VyLySufch/FEaoetydRrnXOP7qAsx57BfWyLoE59K56cXAc+gh+nvyQccrZncMHhnFsRkd8F/kX4M2HXbwVyuUGDBg0aNGjQ4DmBRpnaoEGDBg0aNDjfmCZH72KS2OcfgFdV7sN4QvEHwr47w783BLd48KTZTlApY99UT9gjIi+uHTMK//bYHh9hcg9fGc4zyyRG6W3nQh7JJNP9W2vbXocnK8ETZfPhrx4bs1KnVvd2jYh8ee0cL6kdW9Xv9L3VY5hW19ssvEGlDv5b59xN+BAFT21y3AbsQPWpRKQ99aeBD4X9CT6WKCLyQrziGeDD4d9XA+90zn27c+4zgH8Vtr9YRJZE5Ea8CPmHQ9iG5+PjvWomSbneBDwZiNVngmm7ru7hc0VkMXz+qtr+6et9uojsCZ/rdn0XnghOwvfPCQmxfoLt8WUAInIlPo4sTNrQM8FWtgTws+HfNzIJQ/Dr5+GaDRo0aNCgQYMGlw0aMrVBgwYNGjRocEHhnBsyiZn6LcBhEfmYiBzHJzuqCKifw8d+VMB7ReQuJrExz4bvw7t7LwB3isgdInKUSWxK8KpYgG8XkQ+JyI9tUd6HgF8JX79NRB4EHgFuxCca+r4dlqnCmzgz031FlJ4GYuecVH/A/wn7vlpEEjyB9VjY9vshc/tDTOKk1u/t5SJyu4hU4QQ+AFRq2N8VkXdtUf7Kvf6zReQ+fBzQQzu4t7OpPl+LJ7Hrf1+Dj7FbEX+/F571B/Ek6El84iuAHwdOiciDIvIRJs/lSbx69tOBB0XksIh8FP+c5qp7CsTtF+Bj2p5v/AA+NMAh4OFQb/+huifn3B1Tx7eBB8JxVUzYP3LOPYwnVCuC/u0icgdnt/3vEJF78G1oBm9j//2Z3FBAZUv/TEQ+IiK/Wu0IYTPeh39OLeBDzrm7NzlHgwYNGjRo0KDBJywaMrVBgwYNGjRocMHhnPuv+KRK78eTXTfhXeN/nUAeOueOAp+PJ/Y0nhz6oh2e/714dd6f4ZWJz8Mn/Lm1dti3AxXB9XImyYs2wzcB/x8+BuohPHH0DuCznXN/vZMy1fAmapnuRWSGiTLxbc65cur4Pwn/LgFf6Jw7hVeO/gKTbPVzwF/VfvOTwP/FE6efFO4P59wyXl17H15pqNmY3KrCd+Drrg/MAv+NSbzSs93bOas+nXMpngj9WXzCo5vwRPrvA68McUAJ3z8YyvRJeBf4PwPeGGLUfgxfXxlwczjuY8C/cM79HZ7MXeLc4rnu9B5uBd6AT6Sm8EnK7ge+H/jaTX7yx3iycx6v/vx94BvDue7FJz17BK9QPUlQRW+DL8fHLW4BDwH/j3Pu9u1/siN8H76d5sAnM4llXOFna58bVWqDBg0aNGjQ4DkH8ePQBg0aNGjQoEGDBucbIdP9MeA/Oud+8tkuz/lEUH0ew6swv+3ZLs9mEJGfxif62rMJad3gaUBEPhkfCiMFDgTCvkGDBg0aNGjQ4DmDJgFVgwYNGjRo0KDBhcMS8KPA7z3bBbkAWALeAvzRs12QbXA38G8aIvWZQ0RuxqtuXxs2/XJDpDZo0KBBgwYNnotolKkNGjRo0KBBgwYNGjTYFiLyenyoiwHwl8DXh3jIDRo0aNCgQYMGzyk0ZGqDBg0aNGjQoEGDBg0aNGjQoEGDBg0a7ABNAqoGDRo0aNCgQYMGDRo0aNCgQYMGDRo02AEaMrVBgwYNGjRo0KBBgwYNGjRo0KBBgwYNdoDzSqaKyDUi4kQkCt//WkTevJNjLwZE5LUicl/t+6Mi8lkX6/oNGjxdiMgPishvhc9XiUg/ZFHe9tgGlz+afrXBs4nm+Tb4RMezYeNNu2pQ4VJ/x5+tDJd6eRucP1xuz/5yK2+D84fL7dlfbuVtMMEFVaY6597onPv1C3kNABG5UkSe3EF53u2ce94FKsOrROS9W+xzIjIIBFhfRP7PDs53o4ikdVJMRF4vIrZ2nv5WDW2H5fpFEbkvnPPrznKeloj8ioisichREfmOqf2fKSL3ishQRN4hIlef7R6fixCR7xaRH3sm53DOPe6cm3HOmfNVrp1CRH5BRP7VJtu/TkTMlG2+fgfn+0+hfXxWbduXi8h7gy3dusNybVmvIvJVIvJYaINvFZFd25znC0XkzlD+94rILbV9LRH5KRE5LCKnReTnRCTeSfnOJ5p+9YxjvjbY0L/c5phbQ39a2eZ9U/v/jYg8Evq3D4vIa85yzURETorIzCb7donInwZ7e0xEvmqb8yyIyK+LyPHw94NT+z9NRD4oIusicvvZynUhcD6f705t6hMR2/SdbxaRjwTbe1JE/utOBsib2X0YfNf74FxE7jjLeb5SRH5ni307fq9vZ6sicoWI/HnoO52IXHO2+7uYeDZs/Fkcj74k2Nsw/PuSbc6zZb8pIt8zZWsj8WPJ3ducb8u6ET85fEco172yDbEsIr8WbLt+fV3b3w3v55Misioi79rqXJciLrV3/Nlwsco7jfNl57XfbDbv+nwR+UcRWRE/9/k/IjJ7lvOcFzsPx3+WiHw0jCeeFJEvr+3TIvKjoV9dF5GPicjC2e7zfKKx1Z3hPPbJvyUiR8J44f6p9/9XT/WJw/C+fdk25zsvY9lw/CeLyLvCtY+JyL+t7bsUxrKNre4A58NWxc+XfznYzbqI3CYib6ztr8jgur1+/1nKdb7GD3dNXbcUkbfV9n9G6HPXRORh2WTcPo1PFDf/zwPe/iyX4fOBv9pm/4sDATbjnNty0l/DzwIf2mT74dp5ZnbQ0LYr18eBbwE+uoPy/CBwI3A18Abg34vI5wKIHzz/CfD9wC7gw8Dv7+CclxXk/KwAnc1OLnW8ka3L/74p27x1uxOJyPXAlwFHpnYtAz8N/Pg5lGvTehWRFwC/AHwNsA8YAj+3RXluBH4b+GZgAXgb8Oe15/4fgZcDLwRuAj4Z+L5zKOPlhku+XxWRReB7gLt2cK5vq9nmmMQQkVfgbe1LgXngl4E/lS2U3wGvA25zzvU32fezQI63t68G/neww83wU0AXuAb4VOBrROTrQ7l24W3wv+Ht8b8Cbwv3fLniUrCps+I89fXT2Krv7AL/L7AbeAXwmcB3bXeirew+DL7HfTDwXuAPz1KurfrOHb/Xd2CrFv/cv+QsZflEwKVg41s90wT4M+C3gEXg14E/C9u3wqb9pnPux6Zs7SeAW51zJ7c513Z187vAx4Al4HuBPxKRPduc679OjTfqi8u/iLfZm8O//26b8zyXcSnY6jPB+bRz2HzeNQ/8KHAl3p4O4Pu57XBe7Fz8Yv7vhOPmgRcDH6kd8kPApwGvAubw49z0LGW7XNHYqsd/Aa5xzs0B/xT40Yosdc799lSf/C3Aw2w/xz8vY9kwXng7fr61BNwA/G3Y94k4lt0Oja1CBDwBfDq+7/o+4A/kzIX0hZrN/shZynVe+lXn3AtqbWQ2lPMPwz3GwJ/i7Xge+H+A/yEiL962ZM65bf/wBMJDwDpwN/DFtX0a+EngJL7BfivggCjsvxX4lzs89uuBe8J1Hga+qXad3cBfACt4ouXdgKrt/xPgn4XPj+InIrcDq/jBfzvsez3wZO13jwKfFT7fDDwCfGX4/gXAbeGa7wVedJZ6+ijwyVvsc8ANZ6vr2vFfAfwBnsD8rdr2DeXf4bm2LFftmH8Evu4sxxwGPrv2/UeA3wuf/xXw3tq+HjACnn8uZQ2/PRSe5wngFPCWsF3hG+NjwHHgN4D5sO+aUMdvBh4PNva9Yd+VoSy7atd4aTgmDt+/IdjeaeBvgKunnt23Ag8E+xA8AXIcWAPuAF4Yjm3hbfxx4Bjw80Cndq7F8DsdPv9FuM/T4fPB2rHXAu/Et4e/A95S2ULtfqOzHRv2/yFwFN8e3gW8oLbv8/Dteh14CviubZ7Ni4Dbt9j3dcA/nuOzfnu4/qOEdji1/1/iJ2dnO8+4XjfZ92PA79S+X48fHMxucuy3AX9Z+66C7Xxm+P5h4Mtq+78KeOJcbTz8tulXn2G/Gvb/PH7AOK6TLY7bcj/+ZfnB2vdeqMMrtjnf/wC+Y5PtvWBfN9W2/Sbw41uc5yTwKbXv3wO8u1ZXd00dfz/wL56OzZ2lni/K852yqU/Yvn6T+t2y79zk2O8A3naWY85q96GuDH7itdV5VCj/7k327fi9vlNbxQ+w3XZlulB/z5KNX5RrbnKvm/abwGfj3/NS2/Y48LlbnGdL+5o6TvDvlzef5bhx3UxtvwnIqL2X8e+jb97iPL8G/OgW+56Pb69zF9vGpspxub3jL/nyXig7D/s3nXdtctw/A+64SHb+O8CPbLFvEegD1ze2eumV90Laau245+HFKF++xf53AD9wlnOcr7HsjwG/ucW+8zaWvRyefWOrW573duBLwudr6uXf4e/PS7869dtPD3XWC9/3hXJ1a8d8iDAW2/I8O7jQl+EnKQo/6RwQJpl4Bde9+EnRLnzD3coQznbs5+OJDgk3N6weJn4l5ueBOPy9tnqY4fvJqhLxA9IPhjLvCsb1zWHf69lk8IpXmD0OfEHY/lL8BOoVwYDfHI5tbVFHV0wb2NR+hycjjwZjuGab+p7DdzIH2ZxMzfGTn0fwE73eNufatly147YlU/EvbQfsq237UsKAAvifwP+e+s2dhEZzDg1F49WyP4XvyNvAa8K+bwAeBK4DZkI9/uZUo/wloINfvc2Am8P+fwC+sXad/wb8fPj8ReG8N+Mnet/HxgmkwxOUu8K5Pwe/MryAt9WbmbSHnwL+PBw7i1+J+y+1c30F8Lvh8xJepdMNx/4h8Nbase/Dv+Ra+JXDdbYmU7c8tlZ3s2H/T+NXIat9R4DX1p7zdsTVf6zfz9S+r8P3DSfx9vv9bNNJ4vuVP6u3w02O2SmZOq7XTfb9GfAfprb1gZdtcuy3AX81ZY8p8G/D9w9TG7TgV2odgeg5R1tv+tVn3q9+angmip2RqSfCPb0HeH1t3xy+TVfl+jf4Fc4t+81Q58/bZPtLgeHUtu9iC3IslOdTa9+/FzgdPn8BcPfU8Q8AP3Wu9rYDe7zgz7duU3yC9/Wb1O+Wfecmx76VLSYs52L3wH/iLP0n8Eq8R8Fm+3b8Xt+prfLsk6kXzcYv1jU3uc8t+028QvOvp7b9BfCdW5zrVrboN6eOex3+vTqzTf1vqJupfV8M3DO17S3A/9riXL+GnxQu49vol9T2fS1+4eOnwvXu2MxmL4K9XW7v+Eu6vBfYzrecd21y7E8ThCQXwc4fxgtX7sCP1X+LsFiIb3MrwH/Azy3vB761sdVnv7wX0lbD/p8LZXF44uuMfhfvRWqAa8/y7M/XWPYf8GOG9+LfV28Drgr7zttY9lJ/9o2tbvnc9uHn088P368J5X8KeBL4VTZZ1K/9/rz1q1PH/Qrwa1PbfgdPVmu86v84cGjb8zwNQ74N+KJa4/nm2r7P3sYQtj12k+u8lQmJ8cN4UuQMdSfeJe7va98fBf557ft/ZTKZej1nDl5/KDzI19e2/2+mVgOB+4BP36Ks/wL45W3q7HVAgp+UvQU/Idnqvv8ngfzhTDJ1P3ALvlFei1cZ/sI21922XLXjzkamHgrPql3b9k+AR8PnX2Zq8ocfeG95zi2u8yr8wP2MugH+HviW2vfnAQV+cnZNKF9d2flB4CvC538J/EP4LHhJ9+vC97+mtjoW6nZIUCyF835Gbf9n4Actr2TjSpTgO8nrp+7nkdr33wS+Zot7fwkTEuUqoKRGlOMb9xlk6tmO3eQ6C9QIQPyk7ZvYgYoDv9Lz2i32XRdsUgGfhF+B++4tjp3Fv0ivqbXDZ0Kmblevf8/U6hS+8379Jsc+PzzD1+Pb6/fj3VO/O+z/0WDXe/Bt8QOcRcF4DrZ/G02/uuP+C/+S+zDwyuk62eL4VzBZUHgzfsHhejdpu9+D709KptSim5zreuDBLfa9Fjg6te0bt7Jj/OToT0LZbsCvYGdh3xJ+svSV+IHEm4M9btnnPwP7u+DPt25TfIL39Zvc05Z959Rx3xDqddNB5bnYPZ44/rqzXO9HgO/fYt+O3+s7tVWefTL1otn4xbrmJve5Xb/5/UwRQfjQNj+4xfFb9pub2MqvnaX+N9TN1L6vAd4/te0/b3VOPOG8FOzp80K5Xh32fU+wsR/Ev8c/HU/03nyxbW6qzLdxCb/jL/XyXmA733LeNXXcP8F7Ndy0TbnOp53n+H7hJvzC4h8Dvx32fVWo11/GL/69CP9O/SeNrX7i2mrtGA28Br8oHG9x3lvPco7zOZa9Hz8G+BT84vjPAO8J+y7YWPZSe/aNrW76uxj4v/Xnje/PXo5/h+8D/gj4m23Ocd761doxXbwXy+untn8hXrRYhr9v3O48zrmzx0wVn+TgNvHBt1fw8QKrAPNX4icrFR7b5lTbHisibxSR94vIcrjO59Wu89/wE4O/DcFg/2Ptp5/HmbEdjtY+D/EPbSt8M16dcmtt29XAd1b3HMpzKNzDZtisDGM4597lnMudcyvAv8WTTjdPHxcC+X4WfkV9s/Mcdc7d7ZyzzrlHgH/P9jHIti3XOaCKpTJX2zaHH8BW++fYiPr+neIQ8Jhzrtxk35VstJnHmDTCCls99z8GXiUiV+CJbYuf3IJ/1v+z9pyX8ZPlA7Vzje3WOfcPeEL8Z4Hj4pN4zeEJti7wkdq53h62IyIKPxB7e/jeFZ+Q5DERWcMT4wshTuOVeGJ1MHW/m2HbY8UHqP9xEXkoXOfRsKtqW1+Ct5PHROSdIvKqzS4iPqj98/GrjmfAOfewc+6RYJt34DvvL92izD+IV5o9usX+HWO6XjfBjm3TOXcv/iX/FrwKYDeeFK4CXv9nvGLxNnw9vBVP8hx7GuVu+tVn1q9+C95t+v3blGEM59wHnHPrzrnM+TjT7wnnBz94+HrgBfjJ9z8H/kJEtivXX2+x71z7wm/Hu04/gB/s/C7B3pxzp/Bqyu/A29jn4gckzziw/Ra40M+3/jw/Yfv6aZyt76wd9ya8guCNbuu4kzuye/HJHfbjB6jbYbs2di5958W21aeLi2njF+ua0zgvzxTO2m8CfiyDV9/8+pZ3df7L9VHn3CnnXOmc+yv8hO6fhd0j/Hv5R8O4+514Fc9nn6V85xWX2zv+MijvNM6LPZ1t3lU77pV4ocKXOufuv9DlChgBv+qcu9/5mJY/xqT9jcK/P+ycGznnbgd+j6n2uRNcBs++sdUpOOeMc+4f8Wrqf73JIV/Lzvrk8zWWHQF/6pz7kHMuJcTzFZH58zk+uAyefWOrG8ul8IKnHO/9CYBzru+c+3B4hx8L+z5btk7ud97bEH7MsIwPk1iV9/n4fvRr8fPBF+BzBH3+difalkwVn7n1l/A3ueScW8CrKiUccgQ/qKtw1Tan2/JYEWnhJ0E/iXclX8BXmgCEwdx3Oueuwwdc/g4R+czw82dKGH4zcJWI1F+kTwD/2Tm3UPvrOud+d/rH4oPVfjrePXCncEzqsI7X45U3j4vIUbyk/ktE5KPbnGfTZ/g0y7X5RZw7jX9+L65tfjGT5Bd31feJSA+/4rWTpDB1PIF/FtEm+w7jJxUVKkXmWYmsUP6/xUvsvwq/suJq1/ymqWfdcc7VJ75u6nw/45x7GV4lfBPw/+HVbCN8PNLqPPPOBzgGv1r3mHPuRPj+nXjF1SucDyT+urBd8HW9GOqxfr+b4WzHfhX+JfZZ+GDK19SuQ3jxfRGwF08O/sEW1/kcvOLLbLF/GlvZOPgVpm8Xnxn1KL5f+AMR+Q87PHcd0/U6jWnbvA6vstl0MOyc+yPn3Audc0vAD+Dr60Nh38g5923OuQOhLzoFfMQ5Z8+lwE2/el761c8EvrhmQ58G/HcRecsOy1e3z5cAfxEmLNY593Z8vX7aFr/drm7uByLxycwq1PvKjYVwbtk599XOuf3OuRfg+/MP1va/0zn3Kc65XfjV1+fX9z+LeDrPt15vn8h9/TTO2neKT+b4S8AXhsWorbBTu38z8Cdu86QS1TX34126thpfnNN7/RK21aeLZ2rjF+uaY+yg37wLeJGI1N/NL2LnY7XN3utfjJ+Q3HqW325XN3cB101NpLbsN89Srtu32H/RcLm94y+T8lI7z/m089dzlnmXiLwUH9blG5xzf7/lnXucTzu/nY2266b2TW87Zzu/TJ59Y6tbI8K/l+vXfDWefHsmi6nnNJZle1s9L+ODy+TZN7Y6OZ/glfP78KF2im3ur7KXrXjJCzF+eDPwG7U5Aniy+37n3N+E+eB9wF/iE8huibMpU6tEHCcAxGcZfmFt/x/gSZGD4rOybcd2b3dsgic4TgCliLyR2iqyiHyBiNwQHswqPg6IFZFr8XGj7jnLfWyHdfwqyetE5MfDtl8CvllEXiEePRH5/C0Y89fglSJrm51cRF4gIi8RrxCcAf473s14szL/Ir5TfEn4+3n8Q/yccK43iMjVoUyH8Bmo/2yL+9q2XOF8iYi08Q0uFpF2WEXYDL8BfJ+ILAbm/hvxcavAZz57oYh8STjffwrXvnera2+BD+I7jB8Pdd4OLwXwiq1/JyLXhnr8MeD33ebKps3wO/iVhi8Nnyv8PPDdEjIUisi8iHzZVicRkU8JdhHjXT1TwAZC7ZeAnxKRveHYAyLyOeGnn4d/lhVm8RPyFfGZDn+g2uGcewzvyvlD4Rm9Bi87PwM7OHYWH1PwFF5N9WO1e0lE5KvFrxwWeLn7VsTgdPmn6+WNIrIvfH4+3h1gK9v8THw/8pLwdxgfauBnw+91sKMIUMEO4qdTLrxi5QtF5LWBDPhhPMmw6YqViLwsXH8Pvj3+eWXH4XleGdrfK8M9/sBm5zkLmn71Gfar+Bi9NzOxoQ/jV8K/d/pAEVkQkc8JdhSJyFfjFy8qNfOHgM8XketCuf4Jnji7c5NzdfExK9+xWaGcV4j/CfDD4f5ejV/M+M3NjheR60VkKdjcG/FJf360tv+lIhKLV0T+JD7h2d9sUScXE+f0fDexqU/kvn4aZ+s7PwPfT32Jc+5sk4uv4yx2LyId4MuZvJ+3whuBt08NJOs4p/f62Ww1nKMVvrbC90sZz9TGL/g1N/n92frNW/H9/LeLSEtEKqXIP0wfuIN+s8JmE5Lpc21bN84r/W4DfiBc74vxk7Q/3uJ8XyoiMyKiROSz8d4Efx52vwsfvui7Q7lfDbwBn3DuYuFye8df0uXd5Brnzc45+7zrhXib/zfOubdtc9/n3c7xcQS/PoxNuvi6/ItwrofwXhffG+7xZnz+gL/Yroyb4JJ+9o2tTiAie0XkK0Lfp8OY4yvxoZHqeDPwx1vNc8K5zutYFm+rXyye74jx86N/dM6thuttOz7YIS7pZ9/Y6hn43/jx6hc650b1HWE887zwDl/Ch4W4tbKXqWPPd7+KiBzEjwum1dsfA24Ukc8I463r8TF/N1uk3VCIs8VP+M/4VeeT+CQ372QSwyHCu0acwidE2i4T2dmO/Va88mQF31h/j5CtEx/09lH8hOZJQowvPNv/lqnyPkot/iK1+Ddsnz11Fz4hxo+E75+Ln2Sv4Cd9f8jmgW9/ku0zoH8GPr7VAB/E9q3AjbX938NUQN/Nyh6+fweeiB3i1Qo/s1mZdlKu2vNxU3+vD/u+mlr2PXxD/RU84XaMqex/eOXjvXiC8FaeZkw0/GrLW4OdnAR+JmxX+MncE/gO47eAxbDvmrotTdte+N7BT1Tu2uSaX4MP8L4Wzv8rtX2OWpwRPBF4O15WfhI/CZ4J+9r4if/D4Vz3AN8e9n0YeHntPFeGMvbxK4DfxMb2cB1+oNTHrxK9ha0TUG137Aye1FzHuwV8bXVP+A747fgYUGt4e3/NJvUjeFfFvds8t58MdjEI9//D1OL44FeIvnqL3z7Kxjb7dZxpl7+2xW831OsWx3wVfnI1CHVRz/b918D31L7/Y6irZeAX2BiL9nWhrEN8m970fnZo502/+gz61S36snp7H/erePfrD4XnugK8n1pcMbx9/3CwkXV8u90qBu8X4FWs25VlF74PG4RzflVt32uBfu37l+MXE4b4wcDnTJ3rd/GDn1V8JvAt2+Az+bvQz3cLm/qE7OunrreTvvMdeOVtv/b317X9G/qo7ew+bPtKfF9/tsSTf4R3Wd3umC3f63jS4ed3aquc2ae7C2HLl5iNX9BrbnKPOxn3vRSftGmEVyW/tLZvx/1mOOZAsN1tY7FtVjebHHNNsLER/v1ar7fp8ei7g52thXr6iqlzvQCfmHPAVAbli2hvl9s7/pIt74W0801+94NsnHf9Kp54qPfPZ7xbLoSdh20/hH8Pngj1tTjV/t4eyvQwtQzeja1+4tkqvk9+ZyjPGn4s841T52qH/Z95lmue17Fs2Pav8RzFaXwCqkO1fedlLHspP3saW63b6tWhfCkb+86vDvu/MtzHAD+m+Q1g/0XsV78bePcW5/pyvJhmPdTXT1DLm7DZX5V97LKEiPwVvoLPR1zQp1uGu/ETkrufrTJshku1XM9FiFdsfgw44C7DBicin4pvZ5/6bJeljsu9Xi9VNP3q1hCRnwPudM793LNdlssJl4JNPRu4hPvOCE/yXue28V5psHNcCjZ+Cfebz3rdNJjgcn8ejZ0/d3C51+klbKvNWPY8o7HVC4PLoV43i1d2OeFWtpCoXwyISIJ3b7rUDO+SLNdzGPPAd17mhN8PPNsF2ASfCPV6KeJWmn51K9yGX3FvcG64lWfRpp5lXIp95y68OqEhUs8fbqXpN7fCrTx32/+liFu5TJ9HY+fPOdzKZVqnl7it3kYzlj3fuJXGVi8EbuUSr9fLWpnaoEGDBg0aNGjQoEGDBg0aNGjQoEGDBhcLZ0tA1aBBgwYNGjRo0KBBgwYNGjRo0KBBgwYN2KGb/5UHr3WiFO1Oh9nZWfYsLXHwwAEWFhZI05Sjx49z/OQJVlZW6K/3GY180q6J6lWI4wSUgFMYY8A5wOGcRSmFtdYHcRUBIGm3KPICUcLszDyltczOzdJpt1lZXWM0SrHWjs/hrCOJYoyxdGd6LMwvUpQlaZoyTAeYPCUdriPWIEqBjjClQSuFMxbrBAcorcBNyl6VS2uNKEFrTRzHOOcoS59cWEQQEZRSofzKRwWOFKKEPM8xxmCMRYmmNzeLCKyfXqbIM8ChEBSCiEJHGtGC1oJTGoeMywKglYYqUK5zmLLElGb8vKbVxi784Vwomfj/ib9nxN+DExAl1aOpTjY+3/hYxJeotm9cB1qhlf8TER577EHZiY2dD/zN3/2JExziQEm4D4S8MPSHI06ePMXjTx3n4cePcuz4KUajHMQ/U4LdGeMm9yUS/gARlCiUCJ1Wi6OHD/PIQw+gbEmkNNYarCkxphzbsNYq1KfDWV/vMLEp5xxRFKGUotPpoLTGP2L/G1OWvk0oRdzqcvCqq8hLYc/BKyhMTmTBWaHAkCjDSw7u5wptGJ46Qj9bxxiDQ5FnBWmaMRqlrA2GrA769NOUgY0Ylo7MlDhXUoqmiGdQSYtWLHRmWlxxxQE+/WWfxmte8RoWd+/lbX//d7znfe8DLEaBxZGU3h5cWJrx9wbiIIoVM902e/cscPDAPvYuzTM326PXjogi7W1Had8mAR3akVKKSIX91vcL1TNxwV4tte1TsNb6+85z0ixjNMoYDIasrKyxvLLGiVNrnDq9Rn+YkZeGj/7j3100OwU4evRo4xLQ4Glh//79F9VW19bWzrDVqo97ehB8QlCFOMFhcTrh137773j06DpWII5a/v3qLNaUOCBptZibm2NmZoY4jgFFmmeUpSUrStbW+uRFiTVCUZQMhyP6/T5KKYyxlGVBmqaMRgNwltIUFHlGO4kpTM7a2gomN5h8wPxMzEyvS56PaCVLdDtz6GSVn/qJ72Lf3j0oAbCIs7W6kNq9aXACTnjLL/w6f/Q372FlAIszC7TjHldceRXtTgfnHFmWYkzJ2toa7XaLPM+xOGZmZtizZ4lOu83S7t3jvq7VajE3N8t6f41ut0McRZSFYX5+njiOsdaSpjlaa/LC0O12KYqS0lqWl/tkxQiJYlZWh2BTykKIVcm/+uev4spZKJ0g47FM9bbf+Lxrgfq3xfzczEW1VSYjl+c4LDgDRGEcNBmLAzgRrIA4h7LgpKQUAeso0nX6qyvYLMWYAqcUsQZJ2rS6M8zMzCISIcQYVBgTFggacDhKnFMo0d56HL4tMCmCSP1BufF+S4kTi9iSYtRn2F+mLAZYk+FsidItkAQVdel05+j05rE6AYmIKLDOYiRBO4MuFCYCoyyJxDuptMZWG1wuuKi2urJ8yrnQcL3RKjjjvVCfOLLJ/s2Oee6h4jnS0Yjf+53f5Y477sCWhiSJ6M30mJmdRbRGtEZFEZFSgRvwf1EUjccCWmuUUn4OC+O5k7WWJEkASJIErbX/E0Wr1WJ2fo4XvOAFKKVAJue+ENi1OH9RbbUYrl96BlaRJ36mHDZN3nsEnsBai7WWSGuMOLJRhi1KMlOS2gKLoasVnaRDK2mhtcY5h6q4CmTypq89T8HPxydXDy0xbJDqoOc44u7s066FHcdMrV9hMtAGpbVviDBulGcOsgNJGvZVv5VAeEkgqlBQliVKK4o8x1iLFs3a+hrdTpeyKMiU7ww84QTGlBRFgU403U6HoihwDgaDAQuLi6RZijUlOlLMzHRx1iJak2UFZWFxKD/nqZe5bnGcOXGsH1snE40xgVD0Rq2URmuFjQKR5qz/xzlEFCgFohBsqAvx5Jt4MjCKY1QU45zvJK2zWOOPtcY3OmMMm7WCcQmnnoetiFDAOeufWdiio4gkidHam8WErD6ToAUCQeioaEIlglKCVsrf2gXqnLdC1RmJ85MFHFgHWZ6PycTBcMgoyyjLMjwrHV5IwRYxeK7fjbs8XFVDfkseCDqlNEomtS+iiKJoXG/jMhkbPm9cMLDWUhTFuM0opVE6RmvtO1VjAnE/ucd2u01ZFEikcIFcFzzBHkcRWhyCoAlErlJYA60EBIVSEVEc0c0zciOMSsswzymKlEJiFq64itlde9i9a57de3Zx4KqrecHzbuaKffsp0FUKZt8520A8h7LZOulZEcfGkuY564Mhq+sD5notOu0EE/u68oSzwrqNVuycw7pgrYK/V/F9j5rqRxg/Gxn/1j8/G2yiNvEPfc5knUfhyY8GDRpcXDi/UCiwutrn9PJpisJhBcrCIAJFntNutcA50jRFRGi12ljrxwpOFILQ7w8wxjAYDDGlY329j1J6/I4sy5LhcBje05Y4UiQqCYuZ/lxaa9BAFIHS9IdD4giSJEJrOHRgP3t2LSGIXwSW+jtOJn9OcFJRkI75+VnyLCOOe0Rxwq6lJRBFluas99dIkgRrS9rt1vidoCL/DnYWIp0QqYiiLP3kJ5w5idsY43CuJI40pSlwhL5OOYqyREeaUZb6hTpncc6SZRmYklPLpzDFCJxmaXGOp46e5oq53WEN0V9HxvfW4HLChPDwEybrGU2UkzE9rquJnPNP2gyXefjO93PqqXtIi1V0nIBAr9tBUkUpEZJ0MU5odTos7LqCfQeuB92m3Z4jaSkcgiL242oLogxIGSZ62o9xpRozOpzTWBxODMpqTJazfOQ2Tj7xPtzwNK7MwZVgS99bKIUjQkUtkk4H0+7Qm72Oxb230JrZR9LuUYqlFI2KSjwNcbmnhmjQ4NnFxtnfdu+DzcjTBhXq8xalNb2Znp8XivPvbvDveWe98EkpXCCwlVKTzOFBdOKcw4R5IkyEOlE1fnCOoijGJF2rN0MZxkPgn466gERqg2lsoDKZ8ARq/IzS0ZDV/hrLy8ssnzhJpDSltWRFjhWHVjDbm2VhfoHZmRl6vR4zMzNEUTQm2O0mgrr6vxs/V+MA1wz1ngHOeZRRNWIVlGQCE0XmJqSbjMmLQCo53wHo0BH4RuxQuiKzAglmvdrUuBKlYtI0BfBDozim1+1QFJYo6rG+vk632yHLMjqdLqNRirGWIs8pywIljk4rJtYtT8pojbHrZHmJJZCYEqZ2oWOxeKIK50lD5xziqhUcziBzNig13EThKaKJNEGlUmLDMFIrTZIkYULmlakEklUphYoUSbtFK07Gqj1RfiJYTQ5NaciLQCBqFchNwrL/mS8xqTrjcaEZN6CKCGy1W8RxPF7dOuNdWDt1pYytBsYKvCpVS1gpOTfbeqaoOiNxYPHknjGONMsZDlP6gxH9wYg0zTDGPz+ttVeHak9O+pdTOTVJPkOXE8SqCpxFR9GYnAUJnVpFoNaUvbUXYaXI9tv9BDqK/PO3oY3FUQQCxlaTCGh3Wr6sElSa4QHFUUQSR4jJ/URJZKz4tBG+TRlLrBXtSKOJUTiMi0lLTZYriHtcdd31XHn19VxzzVXs27eH+d1LzMzPg9OkhfF37ryK2vnWE2qE8T2Oa0kE65xXiY1S1tb79Gfa9LodbLuFEkWkozDRDwznVH1V6ulq3qfHTyA8HZHadSfEvydjnVcMW7dBDTz+Gw8imjdIgwbPHhz9/jpFWQJq3P6V0iRxTLvd9gteUYK1Fq29Z0iWl4CQZTl5llMUhuFgSFl6VYYxdjzhKMtyrOhotztk6RDnvGeKX6Dxf06DtZq43UbjMCYlzQYkseY1r34d7STBOb+AOVk0qr0rKnWAc/6cCuZmZjDGkLQTWq02OkroD0fMz84xMztPmg4pjSFNR34xWSlskZNEMZ12JyhPIkS0fx8Yy+rqul/4FEW71cG6gij276E4TlBosixnfbCOKE270wFxdLtd8iIjtTmLi7soi5yTJ0/Q6vRYXhlBlCClqS25Nrg8IX4c4SoLVdWI1msFHDjxo9FSCUW6zgf/4Y957I5/5NAVC8Sxo8w0ne4C+eoaWiXEHUWvo0ErRJWsH19nuPwkhTX0ZnfRnp1jcde1zM4dQGlBRQ5cDBJhSQGLVS1wgd60vjBWLEJJNjjM3R/4K9Ll+2lH65hiQK/dpaJ/jTFEKsJicQbyviXKWqyceJQTj7+Xdu8q5nffyO6rX4j0DmBUjHJlGKM09tygwdOGqB3QopurUOvK+Oc6aVcXmymlSFoJDgvixiI1AT/e0QqFxVnByUYitTpP9TnP8zGhWs0rjTHjz1EUEUURxpoNZan2N7iwqGbJY+uvFKGV56ozDIdDDj/1FIcPP8Xq+hrGGhId0Wm1iXVEN4opnMEoGKUpw9ERTFkiDnbt2sWhQ4eYm5v3vJKajKPHba4SDW7aBhtC9ZninMjUquFWjdA3TjshJGok45hEckwUqapSpFZEKljjtxdF6YmwQKaCXzGxxuIoKYzvBOI4RiuFdY5W0iHPC2ZmZtFK0Wq1GfT74+sba8FZxJUcvOIANig3jHWM0ox+f1Td2ISIrN3ruPzVvauKuAGQsbS+Imr8vkDGOa9KFRGSJMHhVZKVgk5HEfOzs5Qm92XEegWkqDHJ1+m2aSUxrVYLpTXOQV54JW5eFBRFSZal5FER7tWNidSJOo+JerAiux0TBUtoPL5j99fqtFq+zNaN1Y/jeoLJswvkpbMVueiItCbSCq28SvViwpgwOserEcvSUJSW4TBjrT9gZb3PWn9ImuUYZ4l0TBTUuEopyrIMSiahLO2GTqe+Ghi2+Ik9iqTVJs+8G5qf7Kux/L5aAayfx47rdKOrpLUW6yxxUHtXCxDWOYqywFqH1t49Pi/ysRrcOUcrimjpCEoXSE6FVhqJImI0iMKUBmcMzkRoa9BiQEGkIBEFrZj9e5Y4dOggN1x/A0tLi+h2C9HC6dUBRVVsWy08+MgdFsfEFbRq/1X1eUV1lhsGw4z+MGUhKzxBLH5RQSmNDvbkjB2v0FYhFiqCdByWgmoBgHC90CYn5u+JVOcwtvZb6wKx6oImTsJ5mrdHgwbngs08FZ5JO3LOorVmZqZH0mmTxAmt4LLearV9v6sirzw1vscZjTK01qRpRlkaPwERQSvBmJJW0sKUnnDtdXsUZUG73ebE8RPEcUyW+QWwNE3DIBMQwSKMsoLdu+Y5efw0KtHs23uIT3n5i7AmhPapTw4rT4hq7OI7ZqwzRKJYXVmlFbeIkjbGQGks3d4M/YEndI0J7snOkhe5D/vSapPnOVmW4RwUhQ9ZtLCwQL+/jtaahfl5jDX+nWVL+v11nIPhcMjc/DxZnhFFGuMcWiuMKfy4otPB5IbV1RHOwsLSEippc3otZ5AZen71l0ZVdDmj9lLEIk4h4vw400sgwljboV3KbR9+O8ceez9Li5buXBcULEYJp0+tMSxyZhfmSESxsjxAdEK72yaWLuUopd2BVqSwo3WOPXacp1yP+cX99BaWWFjci6gOIl3seGjvxwleMetb0srR+7jn3b9PnD/O6ZMnWNq7j1GWk6YOrGK2N8fJkyssLM7jxIJyPhSWsyRuiBSn0Gt9hutPce+TH2TX9Z/GlVe/Ghd12TCJbdCgwTljg4qtmhbLZK/U92/564kX43ORsamTnwBKCa12gtKKsixwWKwrsc6gVIIO3iHOOR9+kMn8sfKArBaJq/lrkiRjcrVaPDbGjOei1lovlgqh7cZilVo5n3tP5mKgJnQL3z2HYlBKc/zYMR555BFWVlbI8hSxjiSOSTotT7IDxllQgg6jz2qOWxYFTz31FEeOHGH//iu4+eabmZmdhTp3syWJOlW+Bk8b565MpXLn9gpKv8o97bp/5ipUpdRwTsbKVGvNhC13jAkPZ3znUSlFq/NbY1hfX8dZy8LCAqMsY25unsFgQGksyuGJrTwj0pqV1RXKfMShfUtcfXA/eTaiKL0a9fTKOqfFocSFVbeNxOn4fmus/mTbhBD27tkTiba/VwvWUSohji1KRYjSqAgIxKsSWFyYBVtibTEhLYMEL05iOp02nU6HdrtNHCcUxlCUJVmWM0ozijxnFCkynQV3f184G8hS/387VpKOhabWjhWNFfHqFTttup0OCzOzdLtdcC5MXF3tBegbcEX0GWt86AQfOotIK2KtvLu/XNxVr7Iw6EgD4t0lS8sozVgfDFlZ67O6NmAwHJHnZSCuJwsDWglWTV50E/Xu5AWoRHDWoiPt1ZTGoFWEQ4iThDL3RLp/aW0MeTGtUN2w4IAnoivFc1HkqBAXV4Tg/q98LD1rfHxdUYjzLrE4iHVErJR/yME+RWskiom0n7iYosCZEmxEFAI+GIEYwVowYolbMSqJIdaQaCTWKOewzmCdoMRhXTleQfVBKqqFAMYdtgRCwb+wFWVpSbOCwTBllOaUpVd3KaWIAnlsMEHr6mrDr0D8V0ypbOz2x/U3MefxgMOEUBiVq6+xdhweY6cx/xo0aHCh4YjiyIc0UTGtJGF+dra2eKspC8MoTen3B0RRggntuywNDoiiBOccvV6PovAEar/fp91KyPPSK/UBU5Q4W3kF+L7JVYsrwUsmihMsluEopdvr0tYJr371p7Jnad47TtcGpuIq3duYXsXHTPeeGs45Tp06hXOgdUK73SXNC9ZHy/RaHfI8ZzjsE8XKxz4PL2kRGS9Ue3W9oeoNfQzYkjxLEa0ABc7R7XQYDodEUUR/fR1rDIgmHY1otdph4bZFHCXELmZuPmZtZUBucgrrWBsYRoWjF1edbDXoqQ3KG1w28LZaBXaqxb6tXD2Ca+nj936Ih257J+2ZHmnR5c6HVmnN9njtq17GvhtaSOIgG/HYow/x1NETtDuW3UqTj47RimJ01Ga4vE6RCd25BMsp1ldOs3yyxZO6w+69V7O071ri9kxNIVt4zxVi1k49xu3v/g1mh0+hc81cd45BfxmMRVptiswwdIYksqydPkHSjhmlQ2bmZuinlt29BVwG3W5Ema+SyGmeuvsPyU4f5roXfzHEcw1D0KDBM4ZM5pLjbQ45K5nqjxsP0p9jjXGaRK1gnSNOonF1iPJh5+rEqRIhiuJx/FNj/OJplmXj8xRFMQkxGMIgVgRrNc8xxoyJVgkxVqt51nhu1YgSLxJ8O6jmoQ8/8jCPPPqIf47OP8dIKVQU4bRgI0XhHMaBkghtFE6MJ1LDvFZpL25cWTnN7bffznXXXcfirl0kcez5mWl+TmTSHOsz6sYAnjaelpu/BCK1Uj7WFXtbERTVceOOo0Y02UDujd3pw/lMacaqSWu8471zjvX1dfI8J2l16YuwuGuREydOkiQtbHCvts4yO9NjdbnPC2++iX17lyiKnDRNyY3jiSeOBDWcoQpD5pgEcJ6Qk4wnNxV8bEhbI8I2WqBzdjxJM8aiIi/hj2oEmgC7F+fodlpk2RBryhAn02IdtFotWklCu9ul3WmjtHfh82pLQ5qmpGlKrBVpFGGd8cq7cO7q2VhnsUF1a11wGQ9kk4gLZKqfH7bjmF63y8L8HDMzvVB+4wflNVVqRVZV7tPW+g5AHESREGk9FU/u4sBa/ABcwFhDURQMRynr/QErq2ucXl1nMMpCDNNAooakUtVK0Rj1L65SUYZwCyJ02m0cYIIy1xqvGi3L3IdgMN5VY9I2Ji+1uqs51MJkEIh4CCpt64lUpRDrX5ppmqJbMS6Q2jY8v0RrIiQop5yfZGsNSqFU2Ff42GNiNZFoQPlooQLWGO/2rvxkJ7clpVgUFnFCHEeIRFx98ABxrDHB3sYTtPEqpx0TqZMBl+8r8qxkMMwYpjl5YTzvG1TY47ix+MRSlokqurLl6qUwdnOpdKqVATMhU005IVt8uyn9QMTV3f2rR928QRo0uPCYHhtMNBHtdpuiKFhN1yhMQSuOQ7gZjQiUxpDnXp0/HI28+jyQokni++I8z0mShLK0oW8Xut0uo9FptNaUpWF9fS0QlSHWug4SdwtaRZQU4bwgSlMUlszmOEqUBjEVZSphGlm7ldCN1EUAhbEoHdNuz+DQDNOMmZkW3XaL4VqfbrfDaAT9QZ/ZmR79QR8jCpEZklAHxhhGo9GYKC3LElGO9UGfmV6P4Sin227jrE/c5ZzvsyMVUVrvpZCNUlQUiFkn3itDW2ZmZsj7GStra7QloT/M2NXW47jUcBZBQ4NLEg4fCTwShysDxa9ciM2nsAKGguUTj3Hv7f9AbyZmtRBOn1ilf2rEHDGnhnBgfj+tXkzc1bzoihfx8jd0sTbj+LFHOHH8GIkIq6dPceLwUfbN7UblHURS1oaPkSSzxGqBY4+tsHziQXZf+Tz27L0JHSdYIu9xMzzOne/+bVi9n9neLIPckWVDokThrCLrp36SeOokcZyQpimRzEMB6eqIZKbL2vKQditmVGT0R+tg28xoYf2xWzm9dDW7r339s/osGjS4/DERxmxUpW6FxrNhGtPcSBWrNMtS4iieJImq8tCIDxnn516VmvXMZN0VZ1GRpZUCtYqfWbn/G2NQWo/d/yMdYWpLwA0uHDbyR3asHn788cc4cuTIxCu2SjodJ8RRHNqaJVcOayC2lkjioGD2YXqU1mjx3s5KNKPRiDvuvJN9+/Zxw/XXj93+q3IIk3UNcXhBVmMAzxjn6OZPUPGpDQRq5e4+DU8aBqVZjezwCWHMBiJufHwgXCuVZxUryROqBqU9IZVlFqUj0hGMhgm9Xpc8y+h1u5wcDrDGkAksLS5w7TWH6LRi0tGQvNthmJX0eh2fqMhaRKqs4gqtK8WKnBEzdOx6XSN5qu31/ZW2sSxLyHMIcTnHCltrEWDvniV2Ly2QpUPKIqMoCoq89CST8uEBolaLJGmNO0FrLXlR0IoV7USTaEXajkN9TojRMalsDFUEAx8JYKLWE/zqWFmWFKXxIQXiiE67TSvx8WUlEMkiPv5llayrUvtVn8eNVCBSKqh3L24LLcuSiAgRT27meUE6Slnv91kLqtQiLycrhbUJMDCuu2mXBxVUqXV0ul3vnq6CgloLRVYE14rwW5EQesCN7co5xkS91mpMpFZKKU9YG5RUcVd9uIBWEoNzDIcj4k6LKPb7bWD7Y6WJgDIoUwmxABGfCE1FgosLKAt0y+BM6EgRT/KXhtIK4nwKCWccYh1VZGQdxcx0e7zsxZ/Eez7wHh547GHiJMIVPnGW29AjVzrmkGgt1EVpDKM8D+rU7IwQCFprEBtYcZ9t2Dq7YfBga4SqU/gkWOHhTRTYbqxKrfoeH0JhQmL7cwb1aqNQbdDgImHztqZDsqiiMFjbpixLn2yvNDhnGI0y0jQjjiMQS1H42KKFMSTaJ7Zpt9s45+i0Ic9ykjihzAuUUqRp6vuF8N6LooRWq8UoHaKAorSUZU7lVuesQ0tE4RSiNPfe9wBZ/mpaOjnjTsJ6a+W8jAKMA0SR5xmPPXEY4zRJ3A7vF0Oep+RlTrmeUdoS5yxZnlGWBXHcwgGzs/N+kTPEjBURBoMBSZKQ54Vf4HSOdruFiCaO26TpOkmSkOVDlNJkWUmctCjLEIO+KJnpLVBkhiguyHMfmkd0FPpBRTURrjwlmpH2ZYjgnWRxiAZHCVJQOgfSQXCsLB/m/f/39zDpCYpshF0/xuLwMPt7mnbrCPbICsPyk7CLe2m3F3F6NzYyRFHM3iuvZ99Vt+BUAs5R5uu87Vd/hnf8/QPc8uKX0ZmN6c2uoxLBlYIj5ckHP8z68gmuvv5lxN05ynydez/wh3Dq43QMnFwbMioGmMwRmRnKPEdrxfFjJxAR4sh77Jw4fIJur8cwG9LJStqdhJW0YL3fJe4pEtG03CzaHuGh+97D7mtfxyTaeoMGDZ426l6bU/96uKl/60dP/DeeS5ieB1eu96srK9x9592cOnmK+fn58XFRFBPHrSA0icKC8sTTMUmSsdt+NXd0zifoTJJknAcEJqH4dOAPKtFLq9XyY6Gggm1UqRcedb5oOBxy7733sLq6EriY8Dydn09b7fmouCg9wdoCrWNU4Ceq+S0CSlRIoh3iGgdB0VNPPYU1hutvuIFOu+1DO0AQQLlAqFZPfZr1aHCuOEdl6oQ0BMYxOSAo6jYhJcZRUmru5OAJKuP8g7XWerfl8T4fz0nHmjJMmianCa71osjSAUoJZZFjghxkdX2VoixQziGRZvfSErOzsxRFNp58CEIc+07HhIz2lQHaKkP99L1s8g6YdtuedJqTbOK4imj0ZGR1iLOWXq/Lnr17yLMhRe7d9rOsoCyMz9CXxKg4Di8wFzpNRRR5V/pIKxKtKcoWFq9qdaGBuEAkWcdY8TtRlNqgUPEEVF6UjIYpKvKNLc9zdKS9Ard66jWlyvhczqsHbVWH4djMuQ1qy4uFinRX4u3RWEOapQyGIwajEXkRVJvi47l6oq0i9L3KtHQW6wSv61Abn6tzPg6stSHOqt9uQ7IzpfxgoVopbLW73nXTlOOXX1UvFSmoVOjg/MyV8bNTZqxUjkPIi/Fqo0zCSljx6tpOu41WQmkNSqrVSg2iUeLj41jRSJSAOKxnKgHQYjFlRJqBMw6N9rGMnaBQ2JCcpRVFzO7bw403XsN9Tz1I7goS5UleFZKjjPM/V4rucYw2H/84zQoGo5ThMCULsVNhEjR9rIC2noSu7rmqrzpcILlVaF/jpF+BNDGBVPWfg7t/iDHoxgsPLiSFa9CgwcVDnaRzIbliiYgfksRxgojG2pLhcIixhDjUQhxiqaooQhkbtkOW5X4ioS1pWOhL0xRblsRxzGiUMjc3x+rqKoQ+wRhLK0kwykLkyLOcOIqIlCOJEmh12TW/wONPPMVaf8DcQnuKYNw4eay8QpT4uO6jNCcrLTpuM8gKZjodiiInjhNanYSyLOgkHdLRgMFg4O890pjSMByOcM6xuLhIp9Nlfb0fQhD4GKqLi7tYW1uj0+lw8uRpRmlGu9WmPxyxtrpOZ2aGvDSMssIns9AR/f46SlqoKCHP+yRJwozMsLK2iup0EInQqup/Q2++ybiuGhPU9zW96KUDAWILTvmAPs7io0EI4ARtcm5/x9tYfepJ9GxCOVin13+QGTmGihI6zuBOnyBXLbQ9RV+GlMvXs3Tg+exeuAKxCc6NMNqgSSjW++ze0+Kaz30DS0vPQ7cMJ44/zF0fu4vrrr6JfNRHOjFrNufhYsg1N34yTz36QU49+o900wHZCE7nfYg02ekRrZbDaRAUhVXkeUGvl7C2skqr3cYMRz5OYL9PkTtavYROe8mHzSpKCjQqirHFMWy5jE72PMtPpEGDyxnV/G6aTJ2OR7zJu2Kjy9/5L9oljjo/4JxjdXWV++67jzvvvJMnn3yKgwevCsISodOdIU5aOOdDr1k7mU9WIQWdc2NxVuVxV6lSK+/H6rhK3VpPRFVtq+b2zjGZ5ze4IKjzSWVZcs8993Ds2JFx+IayLLC29AIrrUJMW8vJw09hVpbZc+U+WJjDtXqIKaAm6qt7axJ4Jqd8eKyTK8sM7rqDgwcPcsUVV6DFC92UkzBf32zUVpcJNNgpdk6myqQbratShcnDnPYHm8Q73Zx4VEowZaDaFSjt3fkR0JHPdm+NHWc8EzPp0I0xOCyjUZ84jiiCO1u73QbnY1cqhXdHRlGUhtL6DMB56RClcUG5Z/HxHivXtorwqneA1O+TM++nvk2Jd1WuyBogNA7Gx/msekKn00UpRxQpoihG65yiNMEF0RNtxm1U1YmzaOWCMrWFsdH4vFWbrdwgvUrVjSdHrqb0M9Yn8siLEq0UxgDiKKwhCUq+caMNatdJrE87dnGv1D7O+dWUitS62E2xyHOvINU+TIMNMWat8e7yXvGjPelsJ0m0jClxOIoiKBVrha/fg7P+reOzLiqSdhuT5T4xmQhl6d1G/cqit11rvArKrwZOruev7ctQKVWds9W6g48faLVPIiVQZAVxbGi3WmMiuHSOpN3iwL59XLF3N7oYAA4bElCpQNZ6It/HWVU6wokJZGLpk0CE8AxShDAE+MRwXt0iWOcXM8QaOnGHfft3c/3zrmfXwhKnnzzJ4cePMEkSWUsIxeRF7YzDipDmBYPhkMFoSJplmLLE2nisvEJCBsuw2ILbuBBQj5FqQ73XV+Mq5buxG138KyW1NUFRjRurtzcjCxo0aHAhsVHxGEWRj/VlJLjq+1A2WvtxQGn8YhcoojhkKxUhy3KMMWMV62g0oshzWq3WZDEqjidJppx/P+N8WJZet4vJC+IoQTuIlMbalDIfMRwMsS7n1KkVOh1YXetzcHEp9IuM+43JsMcvxCrx72xRitwYhmmOQWGdd8XKsz6lKcD6MVB/MMTYMrxfFa12myj2yRHn5xcYDlPW14eIiigKg1JCkRuyzJOyq2t9tI4oSkeUgIpiWr1ecBETX0+lgdKR5hl5WTDTnWNubobV0+vEccLCwhzdtlAWBbgqE2w9meCZb3Op3fzEI6fBs4mNIgABLGIyTLqGGR5G97pItMSJo8f5wDvfzo3Pu55hukx78AAzxZO42GGjBEfBSHe5Yv8NZKtHmZ1fZPbKm5HeEvmooIXgTIHRKWITjj74cU6fWOf6V7yCVrdL3CrZtXgLy08MuO/OO9h7YIaZXQscO3KKfQcLHh6e5vAjH6F/8gmSVgSSkEhMXjhUrDm1eoooaXk1t0oY5RlZOaIoDU7lFKZAa2h1uiR6gWzU59Tqg+zes8hSdzc6KkhTQ6RShstPMbNv96RWLvIif4MGlz82kqgVkdosoW2Pen9sreWJJ57gtttu48SJE7RbLV74whfywAMP0O12WV5epjQlSmnanQ5a+fwfPrmm29ITuOIlqu3VXKn+vU7EjslUERSTcGkNLhzqHNGxY8c4ceIErVZ7zMeMvTS1QmuhhWCwfOiD72P5/R/j1S9/Obtf/RLUoYN0yzYiE66p4t4cE3GQtZ43M+JYHw646757yG3J1QcPIUzzWFu9DxuvpHPBjsnU8cB5AwvuScKKEd+y2h0+QZEQ3NW9alIphdOTBu5XYEKCqhA7pArGPCbxQrIIFdR5pixZXV2hOzMDOPI8p91OsCYmjjTr/SH94YgoxEZTWqOsHRNi/jZCKOYtJg2eQDszLmz92DoJ6+xETu9/7sBViSUAayitpT8YUZicoigoc+PJH2uwtsQ5KMoyKH6DO74pN2ZoC89EiZd3+xionhwT8YmLrHOoqrxB3l1lM/fqVU9ij0YRzpkxOa6UT3BUTZKqxBiTyVPopMHHEhXBVARteE7TrvEXGqNRhogmacVY48uolaKVJMx2O2gVkRcFWZ6TZT7+rLEWW/j6KEvjbdPBOKFSHb5yPMkswsLiAqeOHfevI+fJ/Tw3GO0ojSOOYpQWTJGGF1jiE1gp7bPyaa/AThKFUg5sgXM+Fp/SEUXuXeGTOA6udDntbsursIyj05vhluffwlX7r2BJW9yRh71LAGHVqWqrCOIskV8xwKJAIoxTOJcTiSKOYiJdYkqDsZ4sd8abrbXGk+fG28fnfeY/4fP/6RfgEB6672F+89d/iwcefDxMwjfQz4wHW6Eyi7xkMByxtj6gPxwyzGaIWzE60iGZl1ecVn18FRfWhhVXKlcVwBmwWocYwP46Po6vo8gL8jynKPzKrTHeNbgoDcaCdVUwfdeQqZcBjIBynkSvTKyK+2O8uaOaBdVLGuPHU71HnMahESytJGJxYZaByWl3Ozjxix22sCCKJNEgflG022n78CpKUFp59XlpiCJNK0n8+6r0iQJbqoNLMx9XyvpF1na7zemVFTqdFtYqiryYLGgJKBWBKNI8RcSSdDVKhf4QCZ7wtSSCU5NKP57xallTGoaDFOV6RMqS5yNMaZjrdMjTguFoRJ5nxEnE2toarbgFIZyPiFAUBq31mBjWiaY0JZ1Oh1OnThMlmm6vR6QjoiRhrd9HxxHD4TD0ucJwmNLrzYQY0haLw5SGsrC0WglZ5jh5fBW9K6YsMiDZQBDvhHyaXlBucOHgqPq+aqWS8Tvfb/X2aEVwGPLBAzz8vj9hdOpRTvTXUDPX8NSRnGtuuoY8H9IbPsXc8DFihoxaXejO027tZWbX9Zzsr7OwsI+4ew1RcgVlPsLHWrfk2TopwvpawcOP3sv1N7wU6xKSlm9DtnS8+vWvQsuncuLEYe6442667RmOP/EIsvsK9u6+Cp0+Dvkx1lYtOprFsUYca9pxOyTkyMlLiwLKvCBLC7JhRpIoWq2Y06fXGXUL5ha7LC3twhQ5g/XTrJpTzM8tIHnB2qlj9PZWsdyrxX4Zz1jG75OL+hQbNLicsPH9VkezNuFRr5dpIZYxhjvuuIO77rqLNE0B7wHa6bRYWFzg2LHjXHPtdbRaLfr9gecTdFCRas9TxHE8Vp4WRQH4bdV1yrKk1WpRhribURSFkIgZnU4nuIf7cU673R6Tqw2eGSbj2lobGbeJ8UsZBYyyjHsffAAiBcqH08uNoawEfUqQSOEURGHePlg5zb3vei97yyHXfcFn0e3u9q79EMRWnhSfDAZ87hHRGieCFocqHY/c/QAtFbH/iivwIwbBICCThfBpFm/aOpqmvjV2RKbWVacVqbpRWhzmRlv83meDD0qH8LAr5YN30ZvEAcnz3E+Gah1GRVQqURgxVKEBAFxwSRv012m327SUwpSekBlmOSIwGKXMz7aC0QhK+WD8nkx1OAzO2PB9k3tj0jHWO8iqburKOf+9iqkqGFeSp6lfAarUrNZgnCNNM1wIXilKAzZMEiOiiKAMDW7ozuGsn0DCJBNcFZqgIrQrWDtxX56oBEP5AEJUBivWq1ZdFV/S/6AKrWCNwY6VxKHzDUxXRc5WdiHOhXJaT8Jd5JY3GKQoHYe69GRyO2kxP9slEkVeGEZZRn8wZF1SBqOc0kxivxprcHWuOhA3XnXrvMN6Rbg7mJtbYOXUaYoyR8QSaYVTGmsJBF5JK4nQUYxQn3gzVsaKUzhTAgZRBlzklbRaEScdnCtDXE/oxhHg3dclbvP8W17Ii264mYV2BxmdZtUZxFlPaKpgD+Ld4J1YBANi0QJOFNZFiMlAKSIdE0cGU5gQGsIitqbANtbXRaS5avFKVKTIyoLu82/kn37B5/CLv/LbrK4NPYFdVVpo81VrcU6gNAxHGav9PqfX15kdzNDqtPygQIVpoIS/UO+Vi36lzh675Vs3Hjj4coYFFuNIs4w8LYIy1W/zf4w/+3Z68Un/BucOG8hU5Tx56qMJB4RuaNIbNbjUIU5wToPEOCkRLJ12C61K2p02pvDJEvK8YGZmJrzjNFFMWMk3SBQ8WSpS3Vj/m6Ikzwuc84u3RWlQOmKm1ebUqZN0Oh16vV4I9i+02h1MXlC6gqxIESeUxhEljjjpIJFGxCBOU4to4++jGveM+zsC4w8ilrgVsbR7L8uP9Jmdi7Ei2ELI0jyMMXzs7NJ6BX2n1cZYQ7frJz8+npZQFDl5ntHrdRmuDWh1WuhY0+52EaXpD4cs7NrFzFyEKMUo8wTycJAy051lNEyDp08opXXMzMyxunKcWCIi6dDrtRiNVhGZnVpfavrHSxcb1SOu1hcq6zBln4fvfBdmcJS1Y0e44/YPUka7yGUf1zzvetRojdUClrmGiIx5F3H10gFiUbSsIOljqIVrGUWOcv1+kBFa5kniXcStFp1YsZwe5+oDt9DrzdLqJUQS+UV/sUgkZEVJZ3YXn/Ipr+KOj3+cnop56Pa7OPi8F5LMXcfx+5/EmZh+ukJ/tIJybXrdeYwdcWp5majVYjhMiaI2Ou5QliWjUUaW5yztWiSKhNGwTxRZhqM+2UyH2YV58sISxZY0Wwtt1lJlcWhsukGDc4Hb4jM0balaQN2o3rWBuBwMh3zkIx/hgQceGJOZxhifRyNz7Nm7l6IsOXz4CPv372dxcTHk2ti4ylOWJVEUjUnS6ns9bFwlRKvc/6MoCqI0vYGrqIjZaT6jwTNDZQFueqvzvM/J48for6/R6XawWMowXwUvavMiwphSCS0n3PKSl3L49juxpzOWP3Yfu6+5lu4r5ildIFErbi7ES7VB4Ka0GoeasoUlKsCWhgfuuZ+sKLj2wCFEa6ySsUe2bHonDXYKdfZDPCrF4iTR0OTzRjf4zX4rNcKjIuY0gvLki67Ir5DZOyhArbV0Op2xC7COdIiJFk1WfkKMUwIRW5Q5ZVEgPtsSaZaSF15tZ614oq22alSRh55k3JhhvUK9s5nueOr3Po5JamxIBuXGEm5rDdYa8twnmrLWK1ErNaxSGq2jEDxaj+OcqBBMugoqHUUROvxbba+CD/vM7+Gvlhmw+qufZ7xdhXgqqsrLPnmA43AHbCSYK0J1Y5xYxqtmVaIsO62ivcAYDkekaUpRFDhriSJFt9tmYX6OpaUFdi8tsGt+ltmZLp1OizgOYRRCmZ2dkM+b6azrMWedc8zNzTM3v0AI74mzjlgrYi1ECsSVFHlKboUcTak06CgoRizG5pR2iHUpFkNhFU5HSJQgkcZYR4midGBFUFFr3GHuWljgxmuvY//u3cx02kTilasmJHQT0TX1dUW0T7r6ILIdd8Qq0uhIU5aFV4Tio8YaZ7GmRCkh0pF3HwmxdSOtmevN8KJbbuGaqw4FctLH0J209+k6hCIvGAwzBoMRw9GIIi/GCaGUMG7/47bFVBu0kwRoRVEpUItAOORkeU5RGEpTUtbscfz8nAtqW7elC2uDSwvahr+aOrVa/a1vb/DsoWrz9b8tjwWcOMAgQf3YardCgkufgEFHkXfftz7ZpCgfviXLC3Skw4KIV3hkWQEIeV763+oY8Mmj2u02cRwzGAyI44Qo8mqO4WhElvqkU/VMqiDEUUSeZaRpSp5nzHS7zM70Jr2oY+oepbZkHPpWB3GsmZnpeldmIAvhBvr9PuloRJamjNIRg0GfOFIQJjpKhWy+ceSTRCj/PhilKVGSMMpyHML6+jplUNsOBgOGgxGrp1cpC8NomHlFflmO1a0q8iEPsjxlMBhQmhJRjtnZFgcOLtJqxWwcVjd94yWNqSZWd+HL06M88vE/gf4TzM7vYnH/QdrdPaytwcF9h5hxQ7QdkdFmzcbEuk3bavLBKrO7l3Bti9YFWd+g8mV0XlLahNSkZMWI3CSYQvj4P74Pa7s4rYlih3U51haIQBFCSWmdECUtbnnhi1jrr7CwAB/74DuhdYiZpU+iTIe0NSzO7EZFMcdOnqDfH9Jp91hf6xPrFqZ0DNYHDNYHpGlBkRtOLa+wurZOvz8kHWUsLS1RFiWrK6vMzc36kEpuEubKI7jJbdQfNGjQ4BzRjJ096kQlTEitY8eP8/d///fcfffdZFk2JkErFEXBaDRi99Ie9u/fT57nDPp9VCC5RCTwCdYn4Q6kaZ0XKYpifF4/DyrGCaeqvzxcuwqJVCXrbHAREKq5KHOefPRROiLokLfGmBJtHQk+9AJxhEQxsSSUKmLvTTdy02tfTTY7g00dJ+95nEFWUGqBSONi7f9VgsF5AZvWoJVfOnQOHbyGXRJRWsfDDz/MY0cPk2I2K+bY6xiaV+S54JyVqXVVKlQTB9m0Ux2r2sJ6sIQBjNbaK1Clcu2PSOIYHUUY00dk4u5fX/GpSMAqQdKE3AqJnVSV4U5RFDZI5TVHjpxgYf4aUBpLCYHA9YZSi8MoEwJ1rHydIlKre60ThfXPnmgUKpd+UYEMspbSlBRlATjvBmVtrT4VWkeAD0SttParDSHmqzNmQ524Kq6ZtThj0FQElC+HsbbWONwGcthahxUDzrvya20QUV6hyyTkgiDjDlusYJwdKwHrwY+nn3lZ+niw07FdLjQGgyFJkpDEMdKKUErotFu04oR2y6sUI60oTekTIQ1TT94xseWKoZnYLjClSh6HpUDYs2cfK8uncCisyXHWEIWwC1VdGmuIVRQSJVV7FMaa8FfinEKpGFP6hChlXiAK4laC1jFONDrWoYyO+ZkZdvVmSIDh+irp2iqmzP2qlNJIFKGiGK28SqSa6leqaaXEZ522CpRDaULc3Mx38tZiqcIglCgtxElEYXxsP+M8+RBpxZ5du7jpxhu47Y67xz2yr6NATG9QcnnVV5pmrK0PGPRHZAsFRVESKQ16arFGxJPCNXur1NiVjVdJpayDMsS9NcaNE8yM46XaiSq+ip9Yf3E0uPRRa5U1dWrzur98UH9WDiRkJFWKvXt38/iJDGP9Qo211oflMDntFpTGMRqOyPMSpTVxK6E0hqL0yQLzvOpHNdZktFodTi0vgyNkuY0pCp/Yqd1qU3YKn6DKOLTSxHGbfmbBmBCD1NJut8jzjIXFA/S6HTZX5bip75P3RafTZs/uBbj/GOIS2q0W6SgnjmO0+GRTw1GOwvmFH+vdkdf6fXrziyTdDqdOnkK3EtKiICpLH5s7ishyn3Br0B8ye+W8r6/Cx5VtJ13yPCdKEhChKDNiHTPTm6EoSlSkOblyklg7ZjszdBGOHzvC4otfFIinBpc2zmBRwyJwRecbHrzv/WRH7qSjFUQ9JE7Yv/86+uka8zMWtXYPvThixhW0WgMinTJinqz3fKSzG5cu073yJXTmb2Jw7G5a3V0oC62eJSss1mQceeJhrj50Jb1FIW4JtnRo7UPo+HjsXiYr2mFcSdxRfNpnvYH77vk4M8eXefC++7jhxuvpnzrMqWNHSDFoq5jtzLC2vkKWQxK3GQxGOKtClCsf6c+WkDsfBqnb6RBFCcPBiDhpceXBK0lHQ587QQx+aO/HtL6FTBZB2PC5QYMG26EhUDeizlFU3ICxlsOHD/POd72T9dW1jcIQ6z1oJHTapjDoRNPtdPzcVYQ8y4ijJCQRDjkyghK1UrdWC6T1ZFTW+hjxVVk8HzJJQqWUYmFhYaxMbeY+5xObjQWh8h5ZH66zvLaM1gpDSV4YSmOJtUbHERJHqFhPvK20phW3eeVnfhbvHow4+cG7KA4f4bqsQHV7XihV90qpiRWlmnwbR1kWmEiQJEZbR0siHn34UbI854Zrr0Mr/16t5tpwpl00Lf7s2DmZWvs87Qa/MYjtxlghG3/vSVXrfDwP8Nm2lXKkeUFH+bho1hqwG0nUulu7CgmGoijBGK/6DFfGlAZrCxCvpCuBe+5/kBtuvIrK7dlJRRJWakPj46SN72/re57cb42EqY3FvEFW5KhGRIXsw4aizMfEplfhxp74EoUohad3IxCLI2Rgt341yomgdE39gyeDfBzLcuJ+7gsMwXW9HhdlEqqg2m6RsOLly2Emx47J4ElMW2fdeAWsOs4/x0m5JorUi9/8hmlKMhyRJDFKtWklPrFRHCuiKLg+lCX9YUwSayLtn4NS+ARKVWdElYhMznjh1F+cDken0+HAoYM89fgjGCy2LCgDSW6cJ7UTV/hkUE6FGCcapyK0RBSmAHx9SnDTN7n/riONyR0qStBxhDWFT1KiPPGpjGWwskz/1EkkHXgyVfChDnRCFF7GJi+C24hPFObJC0E7cKKxYa1EI5g8xZbFmPJ11vkQGFGE0hFZXpCXhqLIMUXO7EyPThJz9dWHUKpSele1tZkN+MFGlpUMBin9YcoozSgKQxJbdK2tVS9/7RwurLRCUHJrjQqdvqjYr7oGhasnU6u/sGIb4iJWIR3shj6rweUAo3w8aHDEge+xUoUTER9T9dkrXoOzIvSb4xhN1i/guRJrYH5+Ft95KEAwIfSOw5GXllFQdUYhA6oxFh1F2JFPQiWiGI5GOCdkuWGU+hjU1pQhTplQBFWriBt7gmitKcrcl1CEJGmRDVJaSYsixKyOo5go0sC0mr2evmHilif4hI2Jjti3dxEthv6gT7s3g9YKZyzWluAs7VZCWeSMioJux8dKn59bxAmUIb56mmU+YV8I1ZMkPv7rwsI8Dlg5fZqZmVmMgSSJSUderZ+XBb3ZGX8fSQuUsHpqmV27luh2umTpgPW1NVq9HkkcMdPpIWJpJlmXA84kAf1js34cMcrpdRZpqRajbMRgfZl8NCLpGEQ/hcqeRBXCuprh7vUeq2Yvexav4mV6P4Ud4PReFrqHOHL4o8zv2s+x5ffQsgL5fuJdtzDIHPfedzdX7FkgiR2RCM4KJowpfQ5L7zLgnEWUX72MbMKVVzyf/PSQ++5/iE5yIyZeQveGdM2IwcqQUbqGijTDYRqEA35M4Mf7fhyjBOJOzHCUY0zJYLDO3n2LzHdmWV1dZXFpgTjSjNZOUpYZEnXRYQ7ia68iOJqxQIMGDZ4ZKiLVOcfd997DBz/4QQaDAViLCh6CNszJTW1+7sVf5YbziAh5kRHpyLtsh3l4WZZjInT6eKUURVGQZdmYUAXGoRSLoqDdbrNnz54tBXANnilkk1BjDlHQH6wzyEd0Z3tgSzAOMQaXeHWp1iqEyPM5WCgtkQOiiBe88Q08ODdD/6HDdLXGWFCRCm8tmXgKE5JSMfFSGYr1ofwwqCShyA1xAU/e/zBRpLn2qmvQwRYq5XMVynDq1hpsgx2RqfV4CsJGF3glVWKmynV4skKzgXhyDmsc1pXjhu9d1UGUX8kujVeY+NMZv0qjJp2QyITYy/M8KOl0WH3xxIpWapyF3onBIaysrPHoo49z7dWHcNZnrh+7sFcMfhhQeTn95lYzrUittlXBEibK1YlLPoBSE/Wcv3dPtkaRT7ijowic4MShJMIZg3M+k59SIfFPrfOru5qPiV7n/IWqcgWXgIqQqggk7zJNcDVQ46x//hlu3ojGxJY7MypEnViuE6rjc11EjEYZUTQkjiMqkXGn0/Kdi9bgLDrEEgmFngyoJSiKagNrN66vibK3Huqgire7tGcPSsMjDz2IcxZblhDq27vll5jCIlHsTUV8AiqcQqsErRKsLVE29XFbcZ4ENQZjBTGgraLT9fscDluWjNbWyEZDXDokLkc4U6KjGKfUmPz0sQmnFzaCnYuglfbJmKwjEkWWjbBF6WV/1uGMJ9C1i8mynMFowNpan3Yr8bFbAS3C3GyPSGtKW7kOyIb6mtSp/8uLkmGaMeiP6PdHzM/kJJEmavu2Ue8jwonGzwSqiZAeK+P94MST4t79d2OYjdKYScKxMJBxQZpqG+LgsoAV70lg1oesHj3Grt1L6LkZbFQp4KdXhhtcsgiDRhEblP8wP99DKe9NYnEgmqSVIEWBc44kaYUwOr5tp3mOTxSosbYgy1KGw4wkCYoOFCLe5W1tdZ2yNMRx4r0BjH+P+5hjXgm7Plgjbsf+XewE60A5QYnmxMmTDIcps0lvQ1bcjeqD2r0BSjQIvOiTbuZtf/1eBrnDlSVFKHeaDkiSCCVQlAU6hNoxYZyQ54VX11uDddb3rcYvEo/SlMXFRQbDlF6vS1GUxHFCno8og/q2KEvvzeAcrbaPxbreH1CYgrzISdptTFkw05uhN98lUdaXmUaZesmjNk5ztf84Z1E4TNYnsstEcUakYtxoSDpcZ72/zqFDBzD2KN3OjZwYKe5fbvPUqmN+qcfePZpZdxc6v5LZKw5y/OjdLM7PsXr0SaKORVozxLsOoLvznH78DtomY9/u6xHVxtrUx4BHBRFAWN1SPi65Kf1YwZl1opalt3+B1mHFYH0F3U7IsiFFf4SRiNKl2KwMCSUNRWHGtlkUuR9H4IixxLHCWs2eQ/tptzoM+kOSzgKjYYqOFKvLx7Bliop6OOeFCTW/uoZKbdCgwY5R91KcVqRaa7n99tv5wEc+RJ7nYQ45CY2HyBnz8uq3Y4/QunALN459WnEPpfHZlUtjvMo1/KbyBK3P9yri1lhLK46ZnZ1lfn7+jHtp8PSxWe258Kwr8ZI4R39lxU/7nSWOIsRZ0DFEETqOiJTyOSGC12SkNGLBqojO7BwvfNUriF6YobsdXBXaUfw41VWvWlUlVa+ShltUXuCcJep1wRmfM8gJzsAjjzzKTG+GvbuWxuTrmFdi4+i20R9tjx2RqRUqxeaGRq8mMRl9hU+pNgM8mWdDIqbgBuQUcdIaS9E9CenjiDgEYx2dVsuTUqak0+mO3XWVilBKk6YjWlFElUyGQHxprcYuvMNRRprmgawNscgiPVYeVp5RE6IM6lJnUWcqUqvOrw5/Xa881FpNwhKUqZ8g4Yk8rWK0ilE6QusYCzjlSVKlgzrXF3WcEGkDgR2I0opQrTrRafVsnWx1gdhzztVigzpEzPgZVvdeJ8XrJOm4YdXOXe+MbYhNWeFix0zNihLXH/gvUoUr0DgvZCIvDHlRkhUFeVFSWoezdVdvhziw/olQFx7X6xKqxQKftEuJZn5pibnlU5w+cRxRPvTCOBlXpCjwMVKcjhBnoEpkElzzvP7Kk4MVkeBQiERYhAghVgoFFMaCE2yek48GkA3QLgVcCA8RgY5QYfHB/1n/3J13+a9YdaUilHh7Vgi2yCjzfFwvxniXlaIoOXXqNP3BAKU01157NbG2WOsVWFprkiSmLELIiMrlsFKnVHAOrE8ANRzlrA1GrK4PWFiYo5VERJEiCq4u9b7GOUdUs3NjDEXl0qAUTvv2Rc3N1i9CQEWyWWswpb8fN05qZcftocGlC+0E4wzFKGXl8Sc5+dBjPFbeza7rrmLfddfQ7s2glF+cal74lwMq4tuFPlDodiOS2LsuRXGMtRlRFJHnOdb6WKL99YHve3HkeUEURaRp6sPmjBdRBGtdUGnkmDIjy/KwsOLDrpRlgYh3qTe2JBL/Hm21WgzXfR+nJKYsLbqdcOzoCZaXV9i/a3bDBGjyXq4R+eKJJBXeJ1fsX+LkySPE7Sv8OMUZ/Di3xFgfu7Usi7BopzGlZTAYMbe4xHq/D0oorGFmfo4sy3BKGPQHdLtdBoMB8/PzZHnG+nqfhYVdDPrD8fs+jmPSLKPVanHixHGysmRudpaiLFBJRCvp4CycPH6MV7/iBjBg9XR/OFHzNbgUUSks8Qu8puTo4YdYPnIbg1MP03JLOJVQSsSnvvrTaPe6xO0b0XYXewcJi0+u8NSJJ1gfPcJ8O6G99xbi7j5W19boLfZYXR2QtCBSe9l39atYd4r1tSP0V5/gyiuvpB0nRNLCifesEedt2xqLw4fxKGxJEntCIDMlhc1BKxZ27WEl9UkwzWAdoxOG2QhtC4wNIamsICgGg/7YU8pYNw7F4ZwjjiNWVtbJshELCz2ccUQ6QkeKdLRGaXMiHNHG0YgnUh3jRKMNGjRosB2mw71V34ui4OMf/zgf+MAHKKtwakGsVc2pnXUoPZlHToRMhH4spigKn5A3zHPqXrl52Fd51Ej41yfaLWvzHhnzEVX8VOccV111FUnw7KnQEKrnBxuW1UWwOBQhF7RzZFlKO4pp6ciH7+t6sZMSv6BfWh/iUUwlThRKBYWDJNco6SJzPayOcFphlecKqkVLn/vGjz0djhLvebzy8BOMVlc4cNO1dHfvxqmIoUCZaLSzPPjAg6gbhb27dzP9Fmz4051jx27+lZy4Ii/q8TCr+IhnNMhApEyTfCKEBEyCjiplpE/ANByNQiwQiKIEHcXe/V0USsdkhaHXm2E0GqEQer0ZT7REmn5/DbCIqVwEPeGlRDMaDceu69a6cRY8xlyqQ1Q9zqfgXDm5/zop6TaSRBPX/bDahHdHtc4RqQnB4zu2CKViRIIqVOlxgh8RwdhKuevLwFTdVajiQE536hXRWnt6k9+Fx+OsC8cZnFNEcVTr1DeqdqsJY0UgVx10XS07WUWbqFi9UHbH+c3OC9Ks8GoeYyrlO9Y6Wq0EELIsY30wYjDMGGXFOGZJpdgdT+5razLT91iHyJiTxInmqquvoxNFrB47TJYNwYWYnmKJohhjC2xhvU2LDQEFwnUdGLTP9iygdITCJ1uLFWgs1x7Yw4qxlDJZOSrLjDIdkqjcu88q5ZWvgXCoSHcbVqlsra0KFkRjsT7BlQjWFBR5jgtxf50TkBhjhDRLKXLL8WMn2bO0h9nZjp/sOO8+6+8nKHhr02/fXELoh8o92zrSLGdtfcDqep+19T7ddkKSxOh4QujX614He6pW5KpFgInN+zIrgvtDiPGrta7ecL48Ic6yL3ujZbwcIM4RIRx98inWnjzCaHmFxT1L6LTgtne9j6jX45Wvey2biOcbXFLwQfFB+8UmKYO6FJJE02op+sMBcavlF0KHw0Co+smCjnzfNhwOfbs3liRJyDLvyt9qtSiKMiSmyxCZvKOTJKEoCpI4IctStIa8yLHOUJgCY72iU0RThjjsDkWStBHpsLq6Pvaisc6BkpA9fWoAaoNKQHzM87nZDtdcdQUr/TbLawNQ1UJygrHlOPxIpDRxFNNqtel2u/R6s96tT0fY0pAXBUVZICIs7trF8soKvU6Hfr9Pb26W0sHq2hoLc/OUxjDXbnFq+RRKa4Zpio/TbUnabUpr6GhFrGJWTp4CNURJQRIL45W+KjV8o967RDGZvk2ekkM0PPboPZj0GId27+OpR0bc9dCDvOy1LyPpzSCiKZyQJo7dkePQ/l2suyWG5YtZXy/pywLJTIuZzgy91gzt7gnWRyPaustwLSMvDSZb5tSpEQevOYRKDJEUWDqIspg8I88yBENejijLjEgLab8ky0YkkUFGGUvWcsXzrmZQxhyTIffmKQ4/9siKAuciTOlj8BtjEaUpjQ/hJTqiMMaHJhLIs4KTp5Y5eHAfWkWcOH6SxaVFHwqAArFFqCuLoM+st4ZNbdCgwQ6wGflYFAW33347H/rQh3zs0rHvnIzHAoL4ZJJMlKxV3NNqHpOHOOgwcbmu3PqVUrRarQ0iqir5bjU/r3KcWGtDnHg99vA7ePAgBw4cOOMeGiL1PMNPPsPyesUXOdLVNXSWk8QatI/3bZ3QEp9PxSqLkohEtUB79SjOkqBx1lEYwWlFIlGwr5A/BDfmiapprnMOjIW85Pa77+Dh97yPl7/oRbz4s99AdPVBrI6gcDilSfOcRx99jE6ny2yvN/bGnYz5GvvYCc4pZmpdmQpBUnzG0W4DeVcRkNUD1jWyrywLXOpXZ9rtDqKEbq+H0oo8S2m3O8zPz3N6ZQUdxaRZTq83492OlWZ2bo5Ou82pUycp8iysWPvstZVC1TkFGB57/HGef+N1RDoOytRAplbKtbE6cSM57KaoljGZKpO6qST2dQWoUoRVojIobxQ6jjyZqiMEfJyUijDSPvCwiEIrsKLATWJE1q8/jXqci+m/8YMgNGzAhGyCFaEaBwWtl6PLBlK0fn6tNXEIdj1+/jIha7e+/sVBaQ1FeLn4cjuKkNBJxGd9Xl0bsN4fMkxz8sK7kfm6gVABgBsrgqV2j5vej3iiH4QkTrjhmmtYuOFa1k88xXB9hSzPKCyM8pzT/QHDIsNhsTpGVIQOKk7vvgouBH5U4tBYYlcQFTk3XXOAl1y7l3fefwR0MlEG4yhMjnGFJ+SV9gQo/pzjbPbBbiMRlFYhzICjtELhHMY/dKwtybLMk8zivKtBrHF4N1uf6CVleXmFubkZSmtI05TTK6u1mKaqRl5WKt8w0WPSvorS0B8MWVldZ2V+hrmZLkms0ZEeB1WvMK3M9m13EgrABNeXahXYOYtzNkzEvFqWsJBSKc8r34VNmlSDSwwWh7KWw488Rn58mUg72vMzrKys8KF3vpv5K6/glk9+MTPzc892UZ+T2Oy9dGa/OTlGnIw5SAn9ZxwrRBlW11ZIWh1EfAKFXbuWKIoiKFA1Ooooy3KcnAEcZejz0zQN7yvIi8wnGBSh1+sxGo0QEdbX10laMXmejjOqKuffjkVREKsQb1prxDqKwkBuePDBR3jly2+ZLF4yNSmZDAogKAZEQaThumuv4n0feow40nRm2qSpJS8cpiiJIkVR+MSBSusQI02zfGqZdrvNYNCn1+uRpiN6MzMMBgPyPCdJEtrtNkprRqOMubk2o3REFGnSbETUikhaLVSUYNKMvMhIki79/pBk1zzOWYx1RJFjbtcMWhcoVXpNhHMTfqny1KFKoOjvthlkP8uoV//4GVkEy8Er93NypYdyMXv2xswtt5lf2IXSLUQUTjSJjiFO6NuCiIJd7RlmE0cky4zSecr2FeSmS9JeYFd7SF5ojq0OWT56lGNP3kOvN8Pe+f1oSowqAJ+AqihyjMlZXz3JcLRKt5ugu220WFy5zsOP3kVUtLCi0eJYnJ1laXeHV3zW67jj3R9jbW2AUzHDwYgk6Y7dGEtbIDpilKUgFhGfUFZEg4rZtbRIu+UXPSwFOBiNUp+4qkwDhTqtS23QoEGDc0N9vGOM4Y477uDDH/5wiM2On2O5Spnqj6t72VWfK69cY8xYRVqf61TzaGOMDwcIKK0x1ngvkhACoEroXf2mKl8lgLrywAFuvvnm7eeyDZ4+zniViH8XO78w3e+v8X9/+/e5+tABrr7pOlw3gdhH8M5cCAMpgUyVLhIpkigiasWoKIZeQhRrCmdxAjrwac46JCz8a0VINg2UFkpLrCLcXItROeTRD32YdrfF8/Z8HlF3npiIDJ+aevn0Ck8++RQ33HC95whkenDR2MvZsGM3f9+oJ7EmgUDA1VzjAzExdp2X2mcqItKMyQtRMiZgvLQ98Z2QCEmrDSKsrfVRotGxIurGDIcjrrrqKk6dWqYsSlbSVYqyHLsiVxntnLHYMODCwfrqWrgPPzJL4gStJm7D1jnEWJyeTBb8hGnSMU2rQOsdku8c/XdjJ52tCSo4EUFFEVJl53OErPchK5/xsWJFRZ7wMYyDVm8WOqHqYKvP1XnGLvfjMASTcprgu62VHt+XUjIOR1C5hVcPqE4Uj78rn/ynDIRsXbFa/flrXnyJmIhXNORZFlw5DWmW02oliAh5UdDvD1hbHzAaZRRlSaUPdXiXb3+iyQupIser828gDYI6yS8M+SxWCsfuuRkOdA+RD2bJ0yE4wYridJrx0OGjPHF8GSPehR8VYS0YkyPOqzudtbi89G6akeKTbria17/yk6EYEIlXPZch5ikC1pRYPBGvRGGdjxvsxvFBwzqWAEohOkIpnyjCOkGiFmJLLFCakmE6orRm7K6glaIsDUncRoknLkejlCIvcWI4fuIk99//QFCPqCBs8qrw8bMJ/x1Pxp1QloZRmrK63md5+TRzvQ6RlpDoZeMzqJTQdRuv4jXralEnjoPthrZnDEmSkBd+oOJjwvhyVGr0OpXe4PxiulbPpNT8t8rNcmwllYlUK64OrIJRf8jKqVPkqysYZWmPdnH7xz9OK4o4eOAAUTJZZGjw7GPzZ+E2fBQ3mQBoBdYUrK2u0eqUtFot8jxnZWVlrDYVhHYI91MWJa1WGxPe+QCtJKHfH5BmI5I4CiSr79/jOGYUPF/Axx7rdrusrq1grcHiSNMMSRTOlBRpysL8At3uDGuj09x7332k2WfRbrU23Mk47rbb6KFRLdRqJXzSi17AXfed4uRqn+FgQBx1UcrHflc6GnvnGGNotdrMzsyCjjFlSStpEUcxRUiwt76+7l3+8pzZQBKrqMX6+jqdTpvlldPs2rUI+JiwwzQlywsWdy2xHsIDeDWLxVphdq5Ht2tw5Ij4ft/i1eAbVbdNP3lJoD4mmWwMKhiDLXNOnjiNkv2M0nVOrD3O7v09Ou0WSITVGbFWdFWHPGqhipRIJ2RGIy4hUglzkmDLk9BuU6xaXNTmtHMcPXWc3fMLnF5tc8WBvbgyR7QgtsQ4S5ZnCAXpaJXl00dZWJhlcXEesZZ0mJIkMyxc82quf+kbILN85B/+mtseeAcHr5vnygM3c+hzDnLPXXfzvts+Qhy3yLOCMrjsOyeMRiMInk8AYv2iRJQJo2FK3muRpSPiGFZWVtm1tIjYFstHn2Lfwg2TeWEliGhM+jmJrcYmDRpsh+k5eFmW3Hfffbzvfe/zIXiCq341HpGxx59B6wiQ4N1SeYdOVKZj1/0wt6/ipFZzeVclNrYWFcIYJkkynhtVxGw1vjHGEMcxV1xxBS960YvodrsTEU6YV9WVsQ2mUO8kpP7V1RaZGcfgrhagvaPDZFwr4njowQc58qHbGT34KINjx9h1YD+tXhcXx9hYY5WPp9+OWxQqgUjQSYSa6WBbLdrtHp2ohVaaobbEhSDGh9nDCUoUEQrrwpy/NH5xP1I8/5bnc+QjH8IcXuaJO+/m+a96JcnVM5RKEZkqgS88+eSTLC4usG/PHj/vqu7WBW+y2hytwZnYGZk69viarKgo8amVxW2cJLvNFJ2VotXhY5vZ4D5fkViW4HanQtyyNtZ55Wa326PV7ozVk51ul9XV0wwGqzhjSNMRohTZaIgxJcbY0OmEMjmfSa8wjrw0IQOaJtItIq09rSJepSdqosasE2dVwqKNytVaoN6KrDReZh+Jv66IV6c48QmBUNpPnJRX1qwNhpRGiJRGaYdFYW2EkxKlHc7pUOeTqrQh3iNK+WD++JWqiihSSoeOehLrdOy2bytZ+GSFzHeo0fgeXOgQnF9WQ4uf4FXn8Gpk/4IwdpIcSylFpLXfHsIPXOwOWovgSsMwL8mdJncpw8IRR7lXQhaGUZoxSkvy3ITJfBUOQkLSEq/KsfU69xUD9X/xZqPDgVZyUJoURaSFOYRRp4WihLJEiWW+0+bK+at5bNcs9zx2lPU8x1iFNQ6xlpY4EhHme112zXTZPT/D1Vfu5cZrDjLT0hQqoaVixAiZdRRR7G3HFqAVWmKcdt5dwKpARpUgOUoZxMUoSRAlWJVj2ovMLl5JqztPng84dfwJTjx5BFMMSSJN6TTK5kCLtLR0WhGtJCIvEtLMcer0Gt3ZiI/ecxcfv/tejA3krnJhvSusnDnw/teCEx8M2LclIcsd64Oc0+sj5tf6tNutoLaKxy69tqpsb+AT2y/L8cS/2q6iiAjfTpJIUcQhGVysiWKFjgUd+UUVEZ8Yo2xipl4QnLmeuTmdqnCUSJWTaNwnW8c4Ci4iHDl8GFVCZi021pw8fIJimLNr3x727t9Hp5M0c+NLGBP/iAgnPsaxQ2PFoAFNRDtOsKVl1B/gjO9FVldH6EiNyZT19SHeFb9gOExRKg4Lt16JPzs3S5L5uGCDwcC/LzGkWeaz6UYaER+DOc8LWq02RZqjJYYIyjKnk7RQDrRTlHmGxC2Onc44uZJyYE+CFgcSUSoLlN4li/D+kDCRcgrlIrSCXktYXz1Mmc9CHDNKV8FYlIY4jkhTQUctRLfJcouOFFEMeS6MRhmdbscnyVpdRSPESiNxgtY+rmsca5zzhHEUxQzWR7SSFkncIpOS2ZkZTp04SZQkPoSSjsiGOUkS057tMdezdFWCOBeSFm7GNDUD6UsCzvrFWAw6hMxwEhTFEnHkyY9Qjp7EFjPkWUE/Vxw4dIjHnzhKka1j8pO0ki67Fq9idnaB2YUFConQSnBEZNagZYQe5QyHa0S6DVJS9ldYefh+TmQZe/ftoZcktNsOUJiijXNDjHOUpwekx59i76427YUFcG2MSTFOME7oWMiLEbv2HOR1X/S1vOddi3zo1j/iZWsPszAX8bxPvprO3jne8X/fQ5pmuAhGWYmyGpEYoyJs6dAYjNKgLYUpOH16mUjDwnyP2YUOURyDCJkbcPTJe7ny+a/DiUaswSnBORWCEimgAJKt67xBgwYN2MgNPPnkk7z73e8eJ5uqCFIFY883FUeIUhTOe9RWhGekI4wpKYpy/D3SPm+GCt59gg9Z6D0Dff9kTUmkEpy1vr8NsVLHsVEJ3ntKcd311/PCF7yAXre74e29WejABmfibOttle+lVEc6iByMkhKLIioi0I6Vhx7BrJxgfXCKB9eP0WnP0pEO3fl52tfsRS3N0bYx/UHKSnaaMh3i4oj24m7mFnczuzRPZ7FDb26OmaSHSnrYRJFbR1ECojDWgbFYW2KkwGmHizX/P3v/GWxZdqZnYs8y2xx/bd6b3lZmVWU5FLxtoA2arofdJGdaQ4ZoJqQIjf5MSArpn34pFIpQxAQ1ox8jieLMcBjRHHGaPd1sB7K7iUYBBVtA+arMqkrv89685thtltGPtfc552YVuqtJoAB03Q+RqHvuPW7vvfZa63u/93vfQ8eP8dSzn+XK8DsMR2O2rt/g4LE1Mu1pmBhsbdRccunSOzSbDTrt9pTUMte8vb8D/DPifTJTZ4ip9wGAmAJlYu45cw8fbv+ur0StDVpXZMIrw3vWE0K306ncvjOajRQPjCdjBJBlk+BkPhphbTFla3gXJhzvarMksefKG2PJ85xmM8U7iJM4tPpLUXciUxv1BFalf9cx1u9dx4yNOvsg5wiubFIiRe02DkpItNKkSUqtlzocjUIy14owZgZKei8r8yIqUMFNz6eUMlRCpueu+i6ed7kDPvxdlZJTlmwNpDof2IRKhjZ/apC4YvvVmi2z43PV62aGVvOgaV3l+qDNpyBMaM6Fa219gbWesihRKkAyM/aynTq4z8DziqxQs45590IjHl54vJ+1RFKBhCKwgWKpcSIBLMaWKCDC0/Dw+OF11noL3Hqwy8bugNwZUpVwaOkA6ytLrC8vsNhukGpFmsTEiUZqjRRNlNK40mBMGYyp4hYiauBFiRYqJFkSHBXLWFQaqtTMZ4eTESrpcuDwaXqrx4jiNh7D4upRovZNmr11Wo1lnIkxeHIzofSCZhpTt8ob6xmPJyTNJvfv3Wfz3n2kC6xyUbW5Tk22mGce7l2enHNkec5wOGF3Z0C31aDdaqK0JhExWu9tfdFKBdC+Gp/1xmFPJVhKIq0xUURsPFFkiOOIJI5Jkhitc+IoojRBpns/fjwxXQuml3u2RswT3qwAWROKxayqq6q5yNXDKS+wzqGiCK01W9vbtNptllYPcOrRR6r5EvZX/A8+/qzC2bsKa3VZ38/zw0N1vTQlUinKMjAwlY7xHnQUzKGctWilMcZWre4JVO6oNk4qYycXuhOsm0q+1HIg9fpVFBlaRzhvkUicCgz+KI4Qpcc6S6PZxFiLEBFZljEajxkOR6j1BZR3WB+Snof599OBPLd9OH3qBEuLPdyuZlzacNzeVclUgRCStNkEodBRjCeYcGWFIctz6Pcr3WuYZBmNRgOpNMPhsOpsCT0Wu7t9Dh08xHAwYDKZ0FSasizRUiJkrU0bhSIiQU4myzNGlLQbS6F4O7053+ua7t9cP+nwEpQDUNXcKZA+wgkPNqd//yq+vIxqCBqp58z6E2iZ0u2Btys4e5QiHzOeDJlMRuzsPGBhZZVmpxc6uURo1/fOIK3HCYPTGW7S54lHjqOj0GYq3BhrFN5rQOFMic1zNjevYe9dY+fGkObJw6yuHyWJOqA0sYCECZsvf5tbjeOcfeZZvvDzv0KqGvzRb/83nDmQsraacfzYUT7zmWf4oz/5LnlmUFKQuQIhg8xQJDTWBd13VekYh/vb0m61WFpcQEpFnudIBYPdLawxyFBJpU5/ay4Ff8b8tR/7sR/7MR9CCO7fv89zzz3HYDCop5RgxlvlW3Wu5CuilmKmZxq8VcLkI1VYmxGhe7QGrWylexpLSWnMNM+JdFQxXSvhEj/zxpBS4p2j1enw9NNPc+rUKdI0neWzD3XTzv93P94dD5FT9/4kQie29ALlwEiwUhBbT2xA2ZADF7bgxrVLGBzFOGOSjygaI04//jTnv/gllo4fxLZjPBqRGYqsz43tm1x96x2uvvI6cliSdhusH1zlwNIqy2uHiI8cJFldwDcb0I4pvaQsHaIQOC/xIkJKgRSelkp45uOf4MbNGzx49U3uX73G4U8/gyYlrw6nBvi3d3d4+513eOrJJ4lVBQ9W42Z/lPzZ8b7AVA+zE1oxxKYt7/BDNyJTEJX5dv8aXameAwgpiCI9bee9c/cOSmmU1qRpysbGZqX76CjzjCLPqmSprHRGw4TkzF66+szsyoHQ3L17j5WVpTDhSUkUzRiZs5FSV2yYtbXXBkx+Rsmfb2ufgZeVi7gXaFQARQnGOVppIqWnTE/vHOPxmKIs8EQoJbFVJWoKPojq8UNM0nkwdAoGVszUYOYRmMECiaB+TgVQKLmndUAQkkypAtvY+pmpVSi07GXfzo5577We/34Pa6l+UFEPMecJrMVqPMmqFd/5WWI/rwlbyx2EBHcG+Myf5/p5wENjzE2RIecdCE8UKVKf4CUIHNYqJA5RlsRKUhYlay3NYrpAebCD84ZWmtJNUtqtJq00JokkcawpjSGWFdygNViLFJKiLABF3OiSdlfQZogY7SIJbE2kwrlQHYOKSUWGVxaiJr0Dp+iunSLpLCJlEvRw28sknaMkSYt2owNWMCly8mJClDbDNbbBpdf4EmMdprAs9hbRUgVmalXCEgJsBVR6AUKFhEcG95UpIA+esjCMJxMGoxHD0YRulpGmSSgQUMmJ+FqLWFatMFQyHWG8We9xQuDljGtircVYTxIbyiTGGEPLpBRFiakquKENx/54BuSHPGQ997/H3+qOBk/YgEQuPN/K0NKP97OxIkB5jyhKxtk43HPjHIRkeXWR5YNrNBa6DwFb+/FTF2FimltrHQ+PjkhHZHmGIKr0rC1xkjAZj8F70jjGGostigo0LVDSk7uCZrNFNimC2Z+xLCz0yIuM3f6APM8RQtBut8kmGZHWTCajwMC3BqVjnDNEOiaDShsctJZMsgmuWv+L0oT2eG8D8ER1THsPcs8jDxw6eJBsMiabBEZprCPyOb0zRyhgqkijo4jSGJQJRU5jDOPxmDzPiaKISGu2t7dZW1un3+/TaDQYjsYsr6zQTFLyPAcCuDTJJqytr3Fvc4PSlHS6XZxzJLHCeUuWGSaTnLEf0vnMOeR8T9G/7/K9T3b5sYZDVvqfjqreXv3Bcf/6q2QP3qKbtPCijUwszqREIkb5CdaWCJ3SkA0akwbOlYyHQ25cv0C3vcD64ZMIlYJQOFGilK0YsDG9hiTL+gx2gvkbQtBXmrTRpNlsh71u5ti4dokHl19idbnJ4MoWd6/d5vGnP4lsaLRtI0pPKjZ5cKfPzvIqrUce5WNf/Bvcv7fBhW/8LioZM7n4KsdOnOTcE6f4/nfeAheMOUtrwBqM93hSpC/RUmONg0SgdDCpSpIEYyxpIwUM/a1Nhv0dFpcOUJ80UXGKqrL6frL4IY39674f7zfqnHA8HvPcc89x79696i8O4Svb3aolugYv687OOI6n7fjOOew8Mck55JycXv1f5wLZyTkXSDoVO7UsgxFf7X8yn88macpHPvIRzp49G77ZFLxliuN86GMOY/mhT5mv/8/4hNOX+qpgPtvWimmnqhcSJzxCOkyZcfeV11kQEqljtp3h+Kc+w6f/k/8Y3ekRLKQshRfIuEGj0+LcgRWePvs07osZr73yKq+99jLfe/FFkpFltbPKytljHD53mpVTJ2isrxF1O7g4phQSI4IxayQkEYHAKA8t89Sv/Dzf2rjH1nhEBjgh0E5gXGUmLiVShSJBXuRErShosAoxNeneB95/eLw/A6qHHvvK4GFqBEPN6Hv4eTUw9W7moqxc7RBBC8SYEinddKJJEnDOcu3qFaI4CaCg9zhTYsocU+RT8ketQSLkXvC2pl57D9ZYdnb7AThTMjBd0rhq9xWzrIdwXHWrOv49WLYEUHKmD7q3La4GYZ0PE2CaNioXdYExxTQxwkeVcHS7Rqynrw2oYPVZcq8Wan18tXtf/ZlTcJvwWqnEdEKfv5KiapMWQmC9mwleh/JVfQL3HO+fNy5mrFr/nqDjBxHOV4xMRJWIuymTGqja4MRM9oAAOk91YusJc64FYo926hx4TfWOswWqYll5hxIiMJGlhCgKzaxSYGUY6zqNkNYS2wC6K6fRWpHEGuUhzy0yamBKifOgLMTCIaWj04h5kHkyY7Fe0uouIZRFZX1KYyE3SBGg9DpVmGrnKoVTmtbSGu31E4hWm1yYqrUzQccp7dU2iVJImTOeTMjKkskkZzFtTidV7wP715SOIis4uLpOq92mvzuqxoQE55DIYKhV07MrFm89QeNCMmitI88KxuOC8Tgnm+TYdjCRC1qKdcGgBj/Do8BmrxYxpabX2gmBk5LYBYOVYMJV4l1dKACECgUbOSEvSvbjJxN18chXU57wVPo8Yf5UAKUlG/TJB0O2t7eYjMf00haLqyt0lpc5fvYRZBzXQo/sp0c/zVFTh2t96srs0QsQkiiKyLPQQaK1ZjLOmYwzBNBKm+DAmSDdo4RGSskky0iSNLBVyxwhg+5yXuRYa2k00vC8ScZ4MkEhAuAaJ2itEEWOsx5jTdCT1hEeKpPKIFdinQv7B4LEDtZMmf/Ci+nO28+Nv6kfqhc00oTHH32EN97eZKM/wphiyky13iKlIklTlI7o9Hps9/vIEawfWK90XfvTNd86R5Ik9Pt9tI4qmZ7Zua0NB0tT0mi0GI1GtNttrDE0W42qgFtJKsUKaz3L7aXgMutzQIZ78t/n6v57vm4/3n/IWg8Fg8Li0OAl2dZVdq5+m8VmgpAN8BqkgcghTQkm7FOLLAM8ykmkbLDQSeimKZPxgKuXXmL96CmS5gJeKqyTCAeiGq/NtEGkJdlkQmEKymKCLydsb9wmxhKrBj7bobO2wplPPIHLSn7zN7/GpdsDfuEXPkM7jSl1hHGOTmPI7s13WFhZI11e4q/+6t+lIScsrHR5+fk/obG1wVNPHmHz/g6X3r6LimPwBaXzGO+JRNCGN84hCe2wSgiKImdj4z4HDx+iLA3tZszu1n12t+6xsLQaNFdFWGdmu/YPXuN/P35ysddI74fF+3nOfnyYwgOFNXztuee4fPkyAEK4aVdU4OTYCmiriEq1Nio1ThLWaGEtQqvQweqDgXLdIRtFEUiBw6Oqx/Xr4soboPYtqb+XjiJ6vR7tdptutxvIJ3O57L6fwN6Yhzumv/jzXlMRAmug1RFgGo8nch7hJFZKMg2osC4V45Ly6m0edTFGw+hAj2c/+wkanSYlHm0sibdI57BSU3iJsI4SR95OOfClT/Llz36EjYuXePOFl7jyygWuf+8Frlx8iyOHj3LmiSc4dO4s0YEldKdNnAQpp7K0CBmR24xR7Dl87Ci/8nf/HlIoZLNLJgVOlUg/53EjBKUpuXL1Kucfe3xaHNgfN39+vC8w1Xk/3Wq8F+Nwepp/yC56avPiZ2Bg/V71O9TOslSbcWMMcRxTFjnOmqrV3Ac9iNKghAxUpoDJMPOa3Xvhw++DdmlopVfgLWmasrjQ48bNO4R+ZKbA2h5guL5pnMPZIPwsHpqg5JyRVcgMw+uCLEBwFReAKXKMKckmWRAK9npaYfJUYDAgrJjiTwJRteftZdwixFQzlSrJmn/OPOuyfixEAK4Cmy84tEu/VwtVSDndZAoxm/znj9d7j7OuAixn17FeNH4SrFQIbRE1IF3LSQRGakhqw/mV0yqTqHbSM22QCmir2HAwf8wVNOndFDgXogZ+KgZd9WZSSpRQKK+QKiaSVK8LWoE4h9IaFXlUBRhJIZAqAqHwIiS4SoUxZIxBK4k3jqNrK9y90ycvPZPckK40kNEy0rTJLOQbGbiyYqISWv8weAWOhLi9zMLBk4h2D6sUUliEcGQmR6kEESk0lkk+YjQZkxmHc3qPGVowmRIY4yhLx2JvkW6vy85gODu3UoRKrQgLjxcB7A6Lu5wC0+G8QFFaxpOM0ThjkuWBPWpsYI/LveNwamNV3x+Ei1kzxutWXl+Zv7npffpuAyslBFle/NjH5ocx6tJOfa1mFSu5Z62QeKwIzGLpA1hgw66SfDji+htvMdzdQQHnHnmEV199lX6/z8LSElEjpbd2IMzy+wv+T13s3YTVkyrTSddXC50niNxHkSKONEUeGKBSaBa6PUajIXle0mw0ABuKuSKs+RLodbuUxtLtdgPIGCmMKYMZRBn+2+0tcPPGDQzgjUerKHRjmAytYwpTUhpLFKc4awNoiqsq9opJljGejANLRFSdJNXgnpPrp6Yr+Oljj7eWz3/2k3zzu/8d1ilQHlHtDSAUOJM0ZXX1EFLHICRShaLuJMvodXsMh0NgliBZ5yizCSsrK0gRoaRCao3DkyQJQgiazQZZWVJaS6vVpD8IcgFKgpaaOE5RQrB2oDXtKKi/+7/P3bRfyvjxR118Ej5GUAQWzLjPzTf/DT7bZCuTTMq7jHdKymyINxtECpTukLYXcT7i2uWbYCZ4p4ijiDwf0GolLC4t8ParF+gtrbJ++ARRHExLShvuodI64qiJ1gk7O1s0GgmT8YhOI2F4b5trm29QJiVPPv0xcuPZHQhOPfU071y+yYs/uMpHPqHIREwjN5hiTLxYkmdDmmUHH1naa4c4ePJRfu8Pv0Z26R6f/WiTL3zuMe7c2WR7KwcJBh12+9IhPRgriCpig/UGITxKK8qyJIlSnAPhHdubtzl55vGqW2fPGf1JXMb9+AlGnZ6ImkH4Zz97+tM+qPDhjHn5vB/84Ae89tqr067H2ghYVnvcGpiab+kP0kVhDhVSouscpMq5a3BUaz0lmE19SOYIPLVhVP3PE/LuNEno9XpT0FVVOq01LlIDYvuA6izqsv4cj+79h2eab0jvUN5N94CuqnVKJwPQ6jw2LzkcRcj1HhtHFxlmO1z9zrcotseI4QgvLGqhxcKRY7QOHyNLFdY4pJWYsaGtYg6fO8/Bs4+y/Qv3+e5X/i0XfvAyd15/hTv37nHq6lWOP3qO5ZPHaKytINtNTKSRXmNQpMaAUSycPI2SMdZDbAoKDJEIxs3z0hO3bt/mzOkzldnqPqD6fuL9aabWxL0KtIPZ4ynl+b120fP5BUxb7ue1PWvau/SzlnnvPUoLGo0EhEcJSaQrEMRZijxHEFiV4TvYGXg4x4adB/9qhkgA2oIu1PLK0rSNP7Te12AqCBG+jzWztnchg67blNkoJDWpUVYgTWDsKqw1OGexNphr1UwcW02CSZKgVWjj9lW7M15MJ0fnZm2E8wCxkAG4q1l69TWpp9t58HT+HM9P7DCb6CFM0DVYJuZAjxobnl7Oufeavk+NjcvZ+/+k3AGL0mBs0C4NRlyuqj7Nm3j5qtOh+oUQ0/NYH7PzM2C+oghTs07dvDMVHi11RdOfVbu11kEjVUR4LxDS44xBVgLRCosUnkiJoNmnJMI5pCtRApQAafPALFaKiRMMM4vz0Ot1Ebe2mWQFW/1d5PGDNNMO0EWhmOS7mN0JMjSOooXHKiiRyKjH4vpZGgsHsXFUsWc1ZVHQH4xotJsoHELa8F+liYTEOlWB8JW4utJoqRFSUxaGOEpodzsYYZFCVaYYdbVrxpaWNXO7AlqdcxUUEkzosrxkOJowGk3I8oKyNKRpXF3P2YZCyFr8YqYjzENjfL6CCx4pIIki4jhDV5qbWqnAssn2wdQfRwScKWxXvHeEPebMfGy2k/Egw91j8WGD6Ty7m5tcfvl1is0dkjSGJKLdaPL0U0/x2quvUWQ5R44dgygsY35/rf/ZiD3K9nViG9iqy4tdxsMhSmucNbRbHUxhSJNGpSsW9MiiKAIrybIMqTUbm5vEcUyehzZ/Ue8lgDzPcc5RFCWdToeyLGdJjhCkjQZ5npMmDYoyI07SoK8oBKaYoJWi3emAm3D91gaffPaJqY55rb09C08w26sXxrBll8Jz6uQRlLD0uh02d3cCQ0AGY0hdGUdZD4vdHpMsQ0eCLCvotLtYawPQG0WUpaUsg4xBUeSMxyParQXGkwlSSzqdFYaDYTijzlKWJUVZMsknIdlSQX+22UgQIqz/ZVFijUHE+8zSn/qoiane4WSMcxOuvfWH3H7nm/S3LV/73hu88fZltu9bPvvsQZ59cpEzj56mu3IEpyOSqMXx0+cQznLr5k3eeP0tOktHWFxcRQF2NOC733iBw8fv8cwnPk6UxCTNBkVZglCUeKSQrKyssXH/Nlpr+js7lMWQlQNLdI4cx9qCye4OQ3WCX/xP/zqDf/JPeP4b3+fo6QarCyv0dzbZHVqWTylGLqOHZmvzGqfOfpSl1UV+/R/+Q/7Jf/l/4cjKJmdOrPCrf/ML/Ma/+BqDYY6QEVZ6MlEQOxdMT33lTyA9jVbCysoKVPvovFSMRhMuvvkyz37qS3ivq9reXIJSsxf240MVPwxI/WGg0z4Y9eGJ9xobVy5f5lvPPx/khZSqiqEzU6l63fcEeb/5Vn2l1NQsKqla/n39Od4TR9F0P6Hq/L36XGPMFEiFmQY8QJIkLCwsVJ00Ah1pkjSd7lHE/or+Q+OHAanvsasL6YqY/XV6XgXVOAj5rkISWxDOUwigKClw7ErLZLjF5TdvM3jnEjaTRCNL23uQJZmC6MAanVOP8OgnP0nr0CF8M8EaR6RUkD+LNb2D63z5f/l3Of+5z/D9b32bN7/7Ande2uLq9SscPXmcE0+cZe3caaJuD5qAErTGEdYHGQCkRAhPnDkiB0Z4pNYVMTGgGf3BgM2tBxw+eHAqgLM/9/3Z8T4NqGDKs5iCk7P2aFe1/T889zzMUJz+PNcu/bAGqbUWvMcUJSM/rDTFcvI8Q0pFs9kiTkx4nrNVK31UJUdiT9UGVzNhfQBrJjmmNOhIEEcRvV6HRiPFmmw6SVljsdruYcLNT2I1+KqUhgoIYg58DWfK4rEBULVzeowVA6fWPonjmCiOK9FoP534ahCipqE/3EK/t9W8+nvV5lz/q7/zPKt03ixrHmCdgsDVMSg1M/2RQuDdDBipnz9tq66lHkQFbql5wP2DncSNKcO4gApIFVPmT0UWCuO1nhB9zR2q/ifqAsFendSa6fjweBaiMvCqmKrVXUEcxyjjkD606XsH3kuEjPCUCB/AykgGhp6zFoFBeYsXntxJPBohm3RWDrK0ssbC6hqNdgfV7PCRLxt+4/f+Hfc27jDMj7PQ6IKMSRYV3ewIZTnCjAaBJSoVaI9XEb21k3QOnEAkTVTkiLxEC8hsxsa9eyy6IEngtafZ0KSNBkJGDIYZRVFSFEUAJrxGRQlChvsuiWJOnzrBqxdeAy/xYqbYK6YLD9PNQ/i/2jAqnJugR2gZjTMGwzGj0YRWq0XThu8zP/6dc2HTMdU+C286D/JDtZkhjNUoijCJQWtZgaiaSGuSOAoGL/vxow8RCl0CgRRu1tkNIORsnhMVbRmCyzKwffcub3z3Bww3H1BOMpIkpru0SCttEEcxa4cOcuOda+g4OLkLVxsZ7bNHfnZiVnCRXuApOXvmBNZmjMd9ms0Ou/0+adJAKoF1lo7sEiUJo3EWin1yNo50FGG9ZzLOUDq0zmmtq7XQ02w2pkz1/vY2nXabpNEgL4pgeKcTZB6MCp2v9MWBJE2wboR1kq9/8wV+5cufR0WCWiKoroWCCDrZ08JyVbgjrEe9dpOVxQ4bwwmRDt/JWlsVfhRSavCCogjdK4mPUFFMlmVYa1FSYsoSY0ww43LBYCtog5ccWFsjbTXC+q0VSRQKUVop8qJACokpDQfWDiBFKPJqlTIZDTl95BhaS/BmH039KQ+HD4VJOcHRQJiSWzdfYXN7hBlIdrZGXLoqeOzRE3zqi09wcn2BlQMnodEidyVCaIwDGWkOnH6EA6fPIYTgxrUrXH3rTVw2ZGW1hVaG1156kceffgbfBGTV8QMYa3CmZGFpgXu3b5HlOUdPH0YKuH1vl8F4SN9IvvCrf4s0Teh0Yz71qaO4UqJG93mwOeHf/v5L/Bcf/SS52WE47PPd5/8df+XX/jOkjjjxyCM88tQXeOnNb6JEwdr6Qf6jX/15/sf/8Y/wpUV4QalBulCkLQqDdQqlBFkx5u79O5w8eRqDISvg9r0NotUdsAVSJdXtKisctZYE248PU8z29LPHCPbkTA93BO7vLT48MT8OALa3t/na175GNslmuTMeKk8Ha+xc51toya+ZonX+orWesvymprq1brpze8ye6/A+5JS19OC89FxSMVKllBhjqs8LOc7Dx7I/fvfGHiB1rmnqPaNOUSoiQL3HEz4Y6Nrpvs9XXiEOJxzSeVx/yGiww8VsgM0dxluWJCQ2JnaCRAtwikJIsp0x/e+8yMsvXubs3/7r9J49j1AahyB2mnKUMW6HL7p+5gy/dvYsX/jSl3j+D/8NV155hTvfuc21yxc5++hZTjz+GI3Tp4h7KzSTRWI00hu8zfHOkGuHMCq0+cOe3NpYy8bGBofXD4bzsj92/tx4fwZUNf2SOabn/CgUtUZlTTTy7z0q/czAqAY85x3p5oHVuopjTIlSOkw4pcVYS9poBV0wZ7AuJCNayErL1aJUcG23lbCur5Ka4WhIURqSNAU8y0tLtFoNJpMAwjlrq+8VgMd6spOVWVVRaSsKIaskUEwX3/nnWufCe81NfvWirZQmTRrESQOtFXEcJs8gGScqndR60IbNK2LGNJ2/JvPapNO/1eCSmGlNTitcdQtf9bi+FsGAqmLJ1ICqDJqf9flz7GXhzC8QU9asUlMgcl7P5YOKGVgf/m9aRKr74vA1Y/1dQLOff/3c+9VA6vz7T6tUBDBQ+JkjtQeU1kgnEa5i8jqFR2BFZRihPNY7SmdRXiCRWK8Zoih9SrN7iDNPf5LVY2eIOwvIKEaoIE+BgGM9zX/+D/4uv/k//za3tzZoLy6QRBFCCZrL62TZmG1zHedHYRyKmM7iIRYOn0I2O3gJsRJo5xESrLCMswGN0TZpBLrVwHpNrCO0jHEuozSGLMuYjCc4mSIxGBsTaQlIfvGLXySOFX/w+3/EJCur1pVwHvcmKn56HWrdWlGtTMY6srxkPMmYZBllxTS2TmJMbX4XqmelsdMxXGsM72FOC1FtbkLRI5jShDEppQos1UiTJvtg6o8rBBbhHViHyXMiFdyVk0YKSiGkwgmBdaHtyXuPdJ7Rzg6vf+8H3LtyjUF/l/5kxMriEt3FBUajEV5Ker0FLhQXeenFF/nS8SPV3TctRe3HT2XU12a2vgVSsgDvUMLTbcesLHe5eWtCUU6IdEKWj4niiEarSVbkaG+xzlIUOb1eD1tpIY/GY0ajESCIZcJoNCKKYwRBWzGKo6ChSgpikcl4Qn8wCBqrURR0lYE0SRmNRpWDbgA9W60W1sZsbo14/c13+MQzj4ZC3fwG84fRHAi6wAu9DucfP8tXv/0SSRIxHE2qPUsoIo9HY3qdVe7f3aDdbGPKnGE2pCgKvPeUZTktXBZFgVKSZrOJc45Wq8loPGR5bYWNjQ0ajSbGGprNJsPxmKhK4lrtNkVhiKKgy6Wkpt1qcf36dZ453cXHf15WsR8/yfDeI6dq6DHKC4Rvsb7+eRhLXvzqN7h1aYNWlHLuyCInDq8irWfr3n26S006nS74FCEjSjcCJdFa4Z3h1PGDrC8klONdLrz2BmU+5tKlN7lz6ybnn/0op0+dCqaOFXnCOof1jrVDh7l54wajQuJtyWBnh7JxmCc++2ssHjjF1v0bPHL+Ue689TzLiwd5+8rr+CxmmJc4ZWlLw/1rb/O5L/01pO4ihEUrxac++wX+q689x0K3ZHHVsboCf/tvf5Hf+a2vMh6NETKmFA7hPIqgCR+nEZ1uk1a7iYo01hoMEV4kbN67w91b11g7dp66uu6FwIu6IL4fH6YIGpRumtfVOYCQAqUEWmkkM1BtH0z48EVNGBqPxzz//PPcvXsX/Cx3DCBm8EbxzqNUwAb25N4V+cg5h68wj/mcsizLKX7wcFdnncvUgKqZM5xqNBokSYL3nqIokFJSFAVpmpKm6Sz/Bdgfuu+KaUPU/OMf8jxf5azhJeFnWWNWTuCFpMBT6uDaEmcZYjThyqtv8MLv/C52tEuJwgtJLBUNL0iVCI1WsaCMNNvSc+aTT7ImY96+cI3vfvWrnB31Wf7IYxQrSzRcSttJtBXkwjMxoSN27egJ/tbf/0dcv3iBF7/+de6+eZE3v/EiWxducOzxcxx85CyTk0dJDhzAtmIi50kz8Eoylo52GUARx0zmUkc6ENMqoEM+3Ka8H++K9wemVm0w3jPVH5wyN0U9xv783bf3hFayuepL7XQHTGnqZZnjK1d5kNUkUSJVRFaU9HoLQWxeCfKioChytFSURY4vQ8twqAA5vHEV0KLIspyyKBA0QECz2WB9bY2tB4EBm9vKfdw7pNRThqVzwXXXmNBCHhglblqdCiBplc4LgTPB8RyCeVRw6QsToo5idJSgo4ikmvSAasKsQNmpKUfVej+nmfrwRBu+7wzkrv9aT8rzGqaiQlSDPuoMHKxBLc9s8lYV0Frrs05dCuvnzLXzP8yarf990FGbgllrp8BKzYz03lXw6uw8zi9o74dFOy9rARWY6oIhiccHnVIPWmmoCgKgQIN3Au8E1kqcBSUkxjusUJRW4NM23SNneOLR86wfP4NudimEwgqJwVeGJw4vJMobepHg57/4Wf7kuW+xeuAwB5dTVKxJuku0spLheMzYGJRUtNtL9NZPI7vL+EijhEMYQ9XjSZQmLCwtkMaSuHIlzApLEkvQCmMscRpTFgWTyQSZRHhRoCaCpK3x3nP8yCH+5l/5q7z64mtcunIjLEBC1JD2HHwyxbrrDvBpGGvJ8pzxJGM8ycjynKIsiaKwOVFKVVUzN2UgS++nmkNTcLWCb60NYKrQukpCwccJSkrKKDDW4kiTZftg6o8ypveVyRkN+jS0RliLA5RzmFGJUwoRRUgdo6IG1jsUknIw5I1vvcDWtVvsbm9zd3uDkSnwwNHJIZTWGGeJdEyz1+XlF1/iyU99gqWDaygPbn/X+NMb0939TCpoZvwSZmwp4PSp41y68iLWWUZ2hJKaKIkY52PanQ4Wh4p0MNTBk+d5VQyUxEmMsY5JkeNEKGy1m222trameqe7u7tEUUSURFgT5EBMUeLsnDlE3X0hwoayLEsePNjkwPI6f/LVr/PM46dJYo3zjll5rT4O8FOToBBSQKQEnU6LtdUVbm5sTdsAtVAkSUqSNDDGEUcJ2TgjimSl817JlygV2v2LAq01zodjbzQaYd9kDJPJJLyn1qgosFm01jzY3qaRpvR3BzRbLXJT0IgjlFJYW+CkRwoNfPBF0P14f1HPq66ixAgfBVNF7Uk6B9jY3iDuaf7BP/xVlpYXOHZwjahdUBQjdjY2eXDxLTrdDsdOPIVIlxBAJAXCOrAWLRQ+6qJaKZ/8/FEGg2280lx+5zK//69+l89+7lN89BMfD3tj73BVob0oLceOP8KdO7d45fWrnP3oJ3j6o1/mwIGnMGWfSFuOnfs4w/4GN7duc/ipL3L1pW/xj/73f4uVo4fYujVkMkloLx4FUqQfE6E4cfwwi4dOcOHWPQ4eKTh6YMK5s6v8vX/0N/n//re/w2RYIJseCxgT9qLNVoPFpQXa7QWct3jvuPdgm4WVdTa3H3D5rTdZP/oYXqi5RLquguyvHT+JeIjCsOdK+Lkn1N1NP6pwzjMcjhiNx0DYOyZJglKK0TBnNB6ytro61Z/mR/z5+/HTHdP51jkuXrzIa6+9VhU/ZyM2sEnjADYpWTFT5UwC4KFuRpiRPmogVUdRyP8BXWuiel/5aKgA1Fas1jpXrTtb61b/er8ihCBNU2ZZ0H68Zzx0an4I9282Dz20VoiKXCg94AQKyUSUmEiSFCU7b77Nc//yX/HN3/8j7MY9ElnQoBU8IaVFlxZlc7wSeK0ppaeUApKY5bNn4Ohh7jz/Kpe/8u/Ymexw4MtfIJcWL1I6JiZTBhlpvBOMjUfFCQefeJJjjz3O3bfe5gdf/RpXL77Nxte+zeEL79B77Cjr58/ROHESFleJZYO20EQUaK+q7uNKjgJFlmWz8+CC4et+ff3PjvfX5j91SX73gubcu82P6ngv5qR1Dj3HcJ1/nTGmmmhUtckPj7MsQypFUeYoHTEejTh77iybDx6glCZJUrRSbG89AKmqpCwAV0JYvAtGD3lRYJxFaYVSEpTi9KmTvP3OVRBgTGDDWuOIY1G1EUJZGuI4QSldMWKD8HRor5sdS5wkKK0pyhKpo8ofS6ClDIYqhMlR6QiJoNNKaTUa4RzVGqR+2vFasVLFnsm3jod/V5/LWdu+eBegWTNW594EOX1uAFC9nDFWp7qwLoB4iJmMgp8mkTMd3JAYhufUi8IHGUkSEU0kpqx0bitw27u5/mJB0Gd8GDz1cxPnQ/vqGRg9A41rBvC0mu2hlJ5YOOKqOimFxogSH3lMGYB96QTWe5wLLrxjUtZPneWpT3yW9vrJSq4BcuOw3oTixZTJCU4mID3KlBxeWQAleOf6LZI4YbHXIEpSOksHMDYnTpukWpO2eujmAi6OscLhrcU6UDp8xyhqcODAIRSSJEnxQjIYjomjJs4LnDckSRSYyzKwqQtrmIxKTNrGJwk4wUpvkS994bNcuvkbeCeQTu1pjWDa+D8Du+sxVLPfS+vJjSUrysqEyuB9EsYeYazZikHuva9Mx2YXTilVsV1FxUyXeBGUWaWUxHFcDQWBTAVaSSL1F1A72Q+q4V8lOn6qqePqm8hbyiyjMGPajRSFoMxyLr95kSRJOHLiBLoVMSkyiuGAdnsB3UjZuHWbSy+9zr0r19nt73B/5wG7wyEGx5bcZXd3l8WlpaCxjaTT63Hl8hUe3LzD8sE1Kt43P5KkeG568GLvZkuwByfbz8H/QiHe9SjMpx4HRFpx8vgazhX0BxmtTovRZIzIIrRWSOnpLi2DFMRxTJbnlCZY0oCntEEjNM9LGs02CEm/36+6DEIRtNvtTlvsjQ8AZDCGCFrbznq8F8EwQivwsNBdIBKO5cVFXnr1Ehv9EevLHXTlKD7tfJg/LqDefId5UPILX/oCf/gnz9FKU0pborRG65Q0bZCmETpWaJkADq0E7VYLYw3j8TjoblfFVWNMaPN3jkhH5GVOt9fFV2tvs9pXhC6bkiSKKIoSZx33h33W1w/RaHboD3c4fmiNlkoJh7rvav6Ti7AGTou+VfHhPfo6QgHfOWy2xc3bL7K9cZFHzh7l8UcPk6KCQZpoYD2kccKxzgFccYLNzTtcvXmZA4ctzfYSUjTwxgAl1lnSdg/lHcPRLrrZ49yTz9JdWODy65f55je+SZ57Pv6xZ1CNGB3HCGOROiUvCtbWD/Dlg4+wePrjmNYyruyjZUGzrdDxcU48+Uu88t2vgCz55F/9ZawvufjyaxR5m6c+/teweJQ0Ac83EbFWPPr4GX77ty5w7V6Phppg8hHdxhH+T/+H/xW/9Vv/hu+99jqJNqCh3W6wsrJEq91FaYXzBaUVvPrGNZYOnufMqVPYIp/qvofz6ffO5fvxwccUMZ3R6KbjXDhwvmL9qenzheBdGul71+S9f5zPReswxtAfDtja2iavNDCbzQaLi4s0Gw2yPOP5F77Jpz7+cVpRp5Ig8igf7kpXsdPelVftx89kTGH8h/Ls69ev842vf52yyPeQNYApMcn5qptVa0wZdFGjSFfdtzNPluln+dAtq3WMr5iuWutZ90m1ya59XGrQdF5Cr2ak1jlNjQHUEkCzHF7s+dy/7AUB72dZQH3G6z18vXf3VTfkbM7wdbNCyO0dSB/wEOo1QvipwfJ0FAiPROGsJNUSMxzw2tee4yv/1f+L/oVLJKUl9gYlBamtzbAd2htaxFgZMRAKcsPRRpt0Y8S98hKttVUOr60wuX6Td/70m3SPHqF19iyjWKONAmvxQmBljBWK3AmcdcRa0XvyMb5w9gRXXnuTH/zRv+W1t99i/cENzJWbHHr8acSTTzBcX0cnmjSRWKUxEqyXeKFw1lFMCnzpkC6cu9pUyzObdwWz81GdpdBhTX0vvTtmbOn3GIM/48Pyfbf5C//uG7E+J/PLYN3i/55aqexZJve05dbvXTt8u8rwxtpKA5VKK9QZnC1568KbJM0mSulKDsAEfUipqDBIpFQIYQGDd4bCULUVqjABOsfBtQOsrS5x6859VBQMoYI2iSVoovmgfepBSIU3odrtvKPRbBE7T29xkWNHj3H6zBmU1ty8dZOtB1ts3ttgd3cHYw3emjAapaCRpOTZhLXVHlpKpiNQuPAFPdPWxJq9AjNG6PRcijkDKjlzSH+YKTpl0NbbdFGbddVi1hXIqyTO+alY9kwDVlW7l9n1nzFd9yZfQR5ABVkA8cGCqa1GQjbJKI2jNOFoZ7yhgFIHkNO/uwV9zxitf5ixbp2bGVcFRmU4lw4CaO4ILWfSEAsLQqGEAjKMl0gtsNkIjyPzgiTpsnroER5//GMcPHmKNNaUPjCRnLWBcUI1+QoQle4nQmOVIFKQGsex9QN848WLxEmT0+IAC+0WabNFb/UwUXMxsJplcIwUziKdxVhXab2C8ppUpdCNMM6DhCLPuXnnBlJLup0e3lviSKK0QscJcaRwrght/5mi220yGuboVsIjZ09y+rETXHjzCsIEoM1POcIhVZ9vqRP4aaXXe7BVC39pfWjzL8tg5OZcdc/6AO47O71krhqTtf7gnpaZSo/YVezqeTZ1Dfgn8X429ReJwKInaAJJgbdhI1knPzafYPp9unHMJB+TJgk6TVl75DSJE0RCcfvePVYPHkBNMiY798g3HG987xWK3TGFydnJR/TzcZiLraNvB+wMB7Q7bdI4wSHoLiyQ6pTduw/wxuEiWRWw/sPjPRM18d5//xnfA3yAIfb+v6i6QITAYZFCI53nwEKKEgV4SVaOQSnKiUF5T39nExlF9JZXMXnQcU4bKXmeEUUR7VabLC/wPqMoDGVRkmhNnudT1mmn0wkSAFFMWRqSNAkyRSbsGaTSJEmD8aRPURRERPQ6C/him1g5doeSyzc3WF/pgrPVGiJA2D3HWm+2w8Kh8UiOHj1IQ1uMz/HCkmcFNtIsLjVod5oMRgM67aChtj3codVuUZQlWZbhnAU8zoauAmcNzWabSZazsLiEdYGpakzQao/iGKkEQmQ0m00meclwOGCx12M0zGi2u0hpiURJW2skHutFYFu8z/igddH/csc8Q9LPfvXQc6R3eGEx5ZjL3/8Dit2XAqNONFA+mEuUJqdwExrK4SaW3XyIFRl5NsaWBXcubbG6dIru2lEiHYGE3HtymxMJaCeabJzTa7ZJj53EljkFGW9eeIXh1lU++/mfI145iJAFqVEkzRiLg6xk4+J32PExB46f58ChVeJ4EaEVh0/GUFq+9f1/x81bB2imHpMPePojn0LhkL7E2wLcCOeCpNeBAx10Aq++dY+GWsBlGcfXx/S33+LX/s4nOf/so/zbP/oOicx47Pwp1taWkCrB2RyhPA+2PFev73Du6XWwOYPBLsZZ4hqXE1Sacfuz+E80PCAqI1evsOHCIKylyDJwCiMVUtcO52L6snqaDb8Qf+6CXOdGk2xCv7/L1s42RRn0LoeTCYUxHFhZIUkT3r5zmeZbmk8//jm8V9Nb04lq7//jOh/78YHHQ1s8AIbDIc899xz9/m6VH4ecAhE8IeRDpBprDaIiJE3ft5INrCXz6hwldLSGjtS6fT+uunankodChk5HYDKZUJZlaOGfA0nLsqzIZ6GLJa7MrX6Y9u9f9vBCInzFHn0In9rLighFcx9AEWrvFFG9QDtP6IAPQofI0M5fE6ug1s1XKGe589ob/ME/+ye88id/TGN3TNs4vHDEMTSNROMpncOqALpLqyFOyaRHxDHLK+t4KUmyEv/WDTaKEfnBDvrGFlf+8HmeWF7DH07JjQl7AF8iqvGjohRvHN4ZRt5g44gTH32WY0+f4ebrb3Dhq9/g0pV7bH7tWyy/fY2DTz9O65HjjJc6lM02LopAB/mJ0hQ45zlyMBCsbDUvKy+wYgaU1sD0dLcyj17/mcNNTH1mpgD2X4J435Ssqb1OTUsSTA2KpqLLFRD4Z76P91WLrpiCevO6lOGmD2LKNRA4D+Za7yiK4L6dFUXYRGqNlIoojnDehcElJVJSbcrcFHhRWiOVRMlwIzQaCcePH+HmnbuhkmQjTGkpCkOcRERJihnn5KWh1WyTtppopTh56hTnzp3DOcfyygqddpd79+9RFAVHjhzh0XOPUmQFmxsb3L13j9t3b7PT30EiyfMMYwuOHT2CUrUW6rsNdGbng/ecGN+r5b/++b1Yq7X2Sv23Grie/0wpZkDqVAdWSaz1WG+n7xeu30zvdv5aBUmED75dsNFIaKQJeWFx3oKrmYtzN7/YW0SeVlD2EoseGsb1wuSnTKp68YQK0K7GaM1yFEIEwwalEDYFN0JKR+Fg+eSTfPRzXyZdWKeIY6yfUNqCwgUhc2Ns9f4hSVeRZjjKuHb1La7evM3EFHzmY89y6uhhnnj8PP/T7/8pr6gEU044evAAS+0ujTQi7bZDVb/SGPTOYUyJMZVpiVBYT9VKosAGDTRrLNtb26wsLtNIUrzQKBUBwXESb9FKolRgt8ZJhC1yBpQcOnSEv/93/hf8t//Db3D10k2EZcr0Vd5PQah5aYrp+SLMBcZYbGkqcNRNCypU59nW7NTq3PvqtfWcMa8tRFHsee8gGl/dHzW7Qe1vif9C4St7MSFwOJQMoL/Ak49GqKKggeTeG+9Q6lApb68sEvdapFKDhAO9daLCksYpA5cjVLgXh8Mh/UGfwXA4XRdMNQ7G4zHOBT3mqBHTajZx1nL79u2qMPQjYqU+FNPq6xy+sS+x96OJh1kTAIcPr7Ow0GB3MMQVgPdY48m8QFhBPJ5g5TbSQqvZxBhDo9FgPJmgdDS3Voa1bTKZBF3wSms3z7PpWhXFEc1GyiTLidMEZxxFbkN3iZT0ej2EMezu7DIYTmi1HQsLi1y9fJXPP/MI3geGACKM/x/qnOuDPnUjTfgbf/2v8k//h99At0KrtVYRSmrKwpImKVmWMRoNiKKI2tCwKGrpo+q8VedrPBmTpulUM20ny1hZWaE0hrI0tDrt6bm1rkQKRVlYnPBMsgGnTx5kZ2uLQ8cWaKQKb+q66V+WLe7PTvg985cjUDQ1eBX+KhwIh/MS6z13rr9Avv0GLQWmLImUocgsu/0R0pVcvHmdQ2vr4DO8LZEqBhmRW8XWgxHf/8FXePojT/DYEx/Byw4ehfAlXgiiKKHVdAy2+mjR4ODRQ0jpuHr1Hqee/gR/+Eff4amPnmf9+ALONnB4dNokjhTSbxPllhe+cY1PfvGvcGDtaWKh8EnGkcc/zpfXT/KdP/0ab1/4Hp/5/C8RNdcYT3aRskTHJTZyGJtjTU6apjTbi9zd2OXetkAWhkY0ptlp4sb3eOpkkyf/81/FWui1HEXepzQZpZlg8gZvvXOXKEnxjOnvbjIqPWWZE0dpdZ4FoN7jauzHBxkhSa+Yn4ATHusMoixQQmIdGBy+sFhl57we5LQ473lvguh7EXuKomA0GlKWhiwvMMaiowjvPTu7u/S6HbTXbA/7PPe9b/KRc58kVqpioQYSwF+k6LQfP/0xL2UHgZzx4osvcvPGjT05+vQfs5y8bsefyt8xY5EqpRAymOwqqfbsT4InjJoSnpxz03Z+IQRSSZyz5JXEWi2FCCHf0VqHfcJcZ2oNttbxcEfrX/aYdjzWuWZVbAlEHqYG1MJ5vBPVCjAzMvV4rHAYHYwehQfhBN46rBRYqTD1R1gos10uv/Ad/uD//o8Zv/YGh33QTTVK4CNFiiN1EolHeYkzCo1k7AzjrI9opui1Fez5I8hmk3RjQLyVc+vODa6llpNHDlBcu09x4TrNpUVsJPFeoa1HmgKERSiBFiDKkA8b5yl8STtq8OjHPsuJ889y9bXXuf79F7l8+Rp3n99i9doVjp1/HLV+ANVtI9KUwjmy8YQjRw6zvH4AI8LZlHMJ0HQEzWioARQV4Tztec78cJtnsQr/7r//jMf7AlMDUCanIERYUMQUVJqGeH9VwcCknAFxD+tu1qBI7VZbt3fUrvfGmEq0WVDmOVEcWo7a7TZSCrJsAiKg9VHiKQtLEI1WSKGIohjvDEqF9vZTJ4/z0suv0x+MKqOawI5TUUSj2WZpZZ0Dq2usHzxInMQgPGfOPEKv15t+152dHXb7fcbjMVEUMRyNaMQNVg6scfbRx+gP+vT7O7z88otcfOMNDh86wPr6en1W3oOdu3cD8HDr/vzz6/PzcFv9w9IAe0sIgVdZX1Mpan1WP9VLrZ0FnfMoAc746SZmHix9uJ1A/QRa/AGajYSsmZIXBo+hMAZhPdaGA1dSTsH00FXupsZp9cmZfm8h9/y+nkgeTjIDtV1UQvXQShJwPmjpeYlSmkgKrIMMRXvtBJ/+5f8Y31pkjMAIS2KDJIX1tmJFC6QMkhKvvvoaL778Cnfvb3Dj5k2eevppdsZjlheXOHbkCIlWaAlXb9xkMhmzubnJ42fOcOLoIXRSmaoYwNcufSXGGiRgrKcwJdJLnJcEvNIjhWRlaZkkjsP5kDEeibUFUoAtQ0tUpCSdTgsda7T17A77rKyu8PixM/zKl3+J//d//88pRnaaDEpcWIQeilmBIPzRWY8pAxgQRNsrIzAxA65NZRQnhKikAOoiDdPfCSGwBAZ2/bH19a31B+t22f14/+GFCMzmakOiRDjH2SRDWMOdazcgyxhubtBMG0wYsHnrDq00JW01WDp2CLnQoowk0ltMUdJodkibCf1Bn62t7en8Uov2Wxyjam6VKjDl4yimt9CjKIsfeaumeKjU+sObVvYzqv/QqItx9c+rKwscXFvg1u0RZemRymN8DlbinWQyyegsLuF8uH/zsiBNE5x1pGkUWjZl0HpWQoLSJHFKluchadaaKI7JJhOUDmBrlCZorxgMRphKI90VBVmW04oTsizjxq3bCJWwuHCE7Y0dsGGzPa3W+9lYmefehyK0Q3qJVpJnn3mK3/nd3+fGgwGRbNJO27RbXbwXOOPQOqHb69Hvb1GMcprNZjDSqZInhKiM+VwwHqz2AJPJhN7iAru7u8RxQhzFDHb7tBpNTFmyduAAN2/codXqYLE0mzHZOCPGc+rk4UoHTPJ+x/Q+4Pqjjrm9hjAVMF+Nn3qC80FTvxiNuXPlBdJoxHYRkzY8L7/2JtJIRjtbEHdZP/cURRxRljnrK4dYXjxC2umy1c+5+tzzXOnfRHz3e4Dh1JOfIPFtIh9TIBmaAi0s7aUOuw8ylvQig3jAp7/8c5z/+b/P8hPf5Z/+X/93fOrZkzxx/klUOSEvBujeMmRQ5kMeP3uUe2+/Qre7QtI9hHDLoASLyy2+9Iu/xNV3DnLo2Hlu3d3kxLFlZG6AXQorKfICa8IeP0kbeF3y2uUNek8e5o0r93ns9BpdrRiPthHpBC9TMjpESlK4Ec47+qOIuzsTjp0+ytHD67xw4zKxbMOcRFUdFT3gA7zW+zEfwoMRggiCoat3vHr9IpuTHZ46fobluIP3QRvXlS7MUtYTJTEh/Xn/1877wOCfTDIKYzDGUloTDFhEDKWnNAaUIDeGKzcvcb+/xbHVQ/jpSPEzMO0v9On78dMaU1yoyh/efvttvvWtb+G9m+vE3EtwmoKs8+8jBEVZorVCaR26atlLOKoB1Ifb8CGwW6MonoKk/cFgqoVe6/fWrf3WBs33ei3WWr8LTIUPGzt1xnqo2/slIKs0z0vAi9CRUDEjnJLIMhgrFtrjpaBUHm1Dt1voeAtkJ5wjlYLR5gavfPu7fPX3fofbL3+fzt1tVo1Fi6pLrupK0pIgV6IhcQopIsbOsqEsuy7n1OIh3PISbm2ZwXafZj5h69YtxDBD5oYHzR0WFpu88c3v8PnTJxkfibAkCOHxqiRSFuEzGl7jraGpG2AssZXEucWPLWmjzYlnP87hjz7J+P4dHrzxNhsXL/PmCy/QSNssrq2RLC3gWw0aS0t88pmPEMdRwOogkLLm4L15WNWL2gVh1nUWCFR78cHZa/0MhnrodT/L8f7FAitKnxDBNTMAb2Lq8jWdYN71sln1pn48L+w8/7t6wpIyVIhc5Xw3bVN3wUSo1h+x1oByZMYQW0vfGOIkDtqf1bfRUYTSGussUZxgnUep4HxeV4KWlxc5dOgg269fQMoIHYVBpKOYT336MzSbHeIkodFsM85HZJMJzXYbRNA5GY/HjMfjytkvJHfbOzskUYySEbdu30UqSRQr1tfXuXblbU6eOEaj0jULyRbT87R3gg0M2nf/fu/5hVlrwcNO9XtYq7UunAtGXdPrJkV17gOQWgO0IVELrpfTtoYqeXuYJTvPiK1B8w8yOq0mRV5SlAYpSyaZoDRgVQXySo2ONFKG4ymNoDY68w/xzeuW/um0MTdup+fT762EC+foNFuh+lUxKmUkSKOCQZExcjGf/NwvQ7OLdQXKh3Z17Ty+8CAd0odWglu3bvPf/bN/hrWew0eP8eyzz/K5z32BKI24t71Lq7fIYDJG64jlxR5Xb29z69598mzE8aNHQSuQEuFqrVuLdwbhqzZB57BCU5QGZRVeaFxV4ZdCcnB9nUaa4KxFJhFZXkAxppkqXAX6KgVxdT699YzHE7JxxnKnzRNnH+HY0UNcvHgD5RQCG1jjiD1jc5497f2s2GKdwxoXWv2NISZ+6DXhPayt55IZS7goTHVdAuhXy4DMXicQKugHOu/2aBntx58fDo+TYfHQ3jPa2iWfTOgt9JA6Zm1tlSIbk+3ssrn1gDOPPsKNW7doLTYZTnIuf+WrnP/8J9FHVoglxDoiL3K6CwsMJyOss3PXK+hsO28ZjkfkeU7abKDjCGccrVabIi+QeH6Ul3EmgVlb8z30N0RVivI/41uAn56Ysiu85fHHTvOt77xN6ZskTU0Ua2yRAynWOobjEUudhamB1NbWFgLJaDig1e5ibZ8sC+t7uxXa+uNqXTemwJiSPJ+w1FoMBpduxkaVWmG9Q0YajAchWegu8MiZs6TNNpMsYzzMsaVFRdMSUGAOImeMiOq4QsEyPEcieeqJxzmwssiDgaEoJVIoTOlJGgmj8YQyt8RxBFUSNx6PKU1JacrQcVGNOSlEkITxlrIs6HZ7oTtCRzjryGyGUgpdtcYWeU6aJIzGI7x0LMk22WjEs0+foduO8X5SsVLfrdG+Hx9QCABTsaSicKUDNSZkgV5g8w3uvvyv2b5+kf5glxe++xL3th3Saf7GF5/kE5/9JdLVw2yNPDujAdaPMWqJfh7zYFRy4Z3rfPu7r9PrrOOSnLffuEQxGfHssz9HKTtY45EOrBRMvCfutHj9pbcxIuXZj/8KXjfodpf43/6f/2+88NXf5lvf+gHHTiyxut5mw0kauovrTygbm0TNLht33uZw7wilahC7ElTOhUuv0GkdJDeSSxdfJVaPc+LIUQo3Cet1ZnBGhAK+CPPC9qDgwo0tnjyxxpuX7rG2HHNkbR2bT2i2FWWWkZsMYjC+zcsX77KVW5569BxvXXwbvCCN02qfXBfIA6kgsFP3Z/KfRAgCWKHwGOEQxnJ76y73zRZvbV3hxuAWXzj+DGvdg6BkZTLsKVyB854kTVBq/t3+7CiKgqwoMMZRFmVYE2zlamEMSkXBWwOFdZZ7OxvsTgYV0aDaP1bFMy/EHpBhP/5yxO3bt/mTf/fHlEUw4qm7NebJTIiQK2kdQUX+msqIKbkHF6mj7rQK+IKe/q6ek4QgdAkSmKfb29tMsoy0MquOomhKVqqZsPOEs0ajMYcrMPuuH6KoGagQWKjzdClHzYIP5MAAeDqMK9EyzEPSeJSToTPUB3NpqwReBmkzURZcfOFFvvLP/jnvPP9d2Nqih6XpoZAGrxWxEiQlKCsQ0iFUjNARznnGJmPXlexKj0g6aN0llj3aDwzpTsmdO3fZnQxIDayXgvv5LhxrwtYuG69eoHnwEziZYoQk0eCkxVPgTI5wJcJ4YhKskOzGith7VGmIvSAWMd3eUY58+hD5R57lwdZ9tq/fo7/VZ/PGFQoB/8nf/wd0um2Es8RIpAlAh52j4tcM/Wkzupid71ow4WGGy8PeE3v+BnPF45/NeH+aqew9xIcPdy9g994nY7o5fw8yTz0B1cwLISTW2WkL/JQ96EErifGBdVY50oTXFllw2vMeHcehgIBHKs3a+iGOHD3C4mKXogz6p0kUBRMg59BRxKlTJ7hw8W3KsqTV6rKwuEBvcZGl5WW0jhmPc0bDEcNxH60VpixRFchbFAXOWpaXluh0OuwO+milg9aJF4zHQces3x+ws73D2sE1Hjl3JrBfnN1zjh4GnwOYWk+072arPvzah1k+8yzXvQCtRIjQjiCEDLqp1YIgKwbsvN6qECBkpSkiCAZezNix8/9qEPyDTsY67SZlWWKsQ6kcJQV5KUILshBoFRzcEYKytEARQMWHzUOECKzVuV7e9yoG1ICdr//uHe20gXcuXFevEB5skeG9YGH9CFFvicLlKFugnULjkN4H/UlUcKBWmm8+/zzeOc6fP8/pM4+gIs3S4hIbD7bY3dlle3ubR44fpNuNWFta4OKVmziraTSWWFhcINIa74JZmi1LbFkiXHDL9gSTM+s8zli8siitEQKsdSghaTQaSCEwpSXRmt3+Li4bEq8sTI9fVou6957S2SBcnReoRcXyQpczp0/yxtvXQlXPumoyrqZaH1rDnZsDo7zHCzc1gbOVmYwxoXgitUIIPze2goTBPGO71g+ux64xbmpaVR+3977akAczup/lCfwnFRIfDFBGE3bu3WdlZRnlPChodTssLPfIioJDRw/z5rde4NT5R1k4cpg3XnyZA4cOcvviJZKNDQ6dPEraThmNxljvmBRZBf6HsVQUBdZZnPds7+xgjKHdbkMcY3xBnEQ4POPhiGbS+9Ed4HxhZSbIxrQaSwWS/eg+8UMd84UVZw0feeZR2s0/ZpCFTa/SAm8V3gdd79FoSKpjvHPEUUwcJWxvbyOEYGFphbIs6XRaDIcjIMzbCwsLlKZgPC6qNUwxGo1YWOhR5nnQHFOSONEYU1KWoeuk1WqT57s82NwmaeYsdNcxVoRipBRVal2x3fZU5MU8Kh8KdFgiHfP4o2d56/K9ajYUDEcTOgtLmNKitcTYMmw+XZC4cNZgrcG7mtkSEqlGo0Gr1UJKwWQSgKjl1VVGoxGdTid04VRzZRQlaDVhlI+QSqOkJtGOpYVm0Jl3kqD7qveB1J9ETLd8Du91SOSEx2PB1+3MjmtvfZ13Xv8GG9dG/PZXvkOZDXjmqWf5a3/nSxxYTBgUCd97/Qo+syRxk3HWx1oo83uk0QJf+Z3/HydPrHHkyDHa3Q6LfpPbF76Kd4Zzz3yRVrxAhKP0Md5rVOQ59cyjDDlOe/EUt99+lXdeeZVjj5/Gdg7yy7/0n/HOK88hGznbm29SNBfY3i25/sarfP5v/h1arQ7KlDjdwnvHy9//Ommrw+Hjp/jjP/gdzp47xttvvo6KI9YPrGKyMaYw5HlJvz/AlTmunGCs58bdPovdHkeW17h+/w63t25w9PAicT+jkTRoNduMJymXbvW5389BNem0lhnc28KUljhKqPqImM7nYv7k78ePM37ovCJACk/uSq5u3eLilTcxDUujLdGx4Ae3X+N8bjmxdjx4Mfjax8ACjjhJ0FpVHQLv8bnVf42xjLOcbJKT5TmTLA/gFmFnapxFWxnMNH1gFI7LMQ+GG8AZ8FVpVczS//2R87MXD+fP8+NyMpnwgxe/z/b2VqhjzXWB1k+TUk6lB2vpMKlm/iZ1O36dC8MM51BKTI2FlVQVYQyCqXP4PGtL7t3bIKuA1Pq1Wuvp+8134tV/a7VaU6bqwx2uH5aYN0WCmjnpQ4u+d1gEhTMop3hw4zpvfes7XP3+K9hYYoCoFDz60Y+wcHiN8OwAmo8GA37wnW8zuH+Pi9/9HmZjgy6CpnckCgwWFwdDY2klkRAIZ1FovHPkROyInAflGOPAFZLlpImRmpXeAtHmkGhzh3zjAbvaEBnJYibYjAuuDDd4cvEQl37wEo98/BH8cor2mjgT6FiRScPNi6/Rv3qdxz/+SaKDx8kkSOOwInS2KCdQQuFkxMgV+GaHhV6b5ROn8caT5wWD0ZiFQ+sUEnRFOJTCo0TYi4RW/fo+8NNu0dqPJPgfMANS666th3KlPY+8n3v8swuovi8wdQrwwdR1HiompNg7iVRnuzrhs1Fda4zMM9EguNUhZslUXWWpWY3zLMvqnWg2mwEgKQuKLA8TkfMYSkxp0GVJu93lxPHjnDx5gqPHjnHj+nVu377FaDTkmaefRGHwPohJCyk4fvQIy0sL7OwMOHRwnVa3S14U7O72WV5eIS8ykiRFeE+R5exsbdFoNLDWTrVMOp0O3nsazWaoJCUpjUaD4XDCYDRm0N9hMNim2WiwvLz0EOD4MOBJBXiGSmgdNZPyz6o27WX6PiwbEN4vTLbBrKsWc6/Zv4G9Y6ZM2zrCzVJVe6ag7V6wdv57KfXBalG1280KeINIBTfYLNdQAZS1FqxzjklWVaRLG4wTYDpZ1IZnsxBT1uTeqMXHw5ZKA71OC+FcaA2p5OkLIzFOsbZyGKEjcAXSW7zXOCROlgGZtQItA7j36KOP8ZFnP8rtu3fo727T7Xb5xjee4+79TayMuHvvDr/2138BrSQLvTbFpI9OEo4eXKXTSXFliS0LjCnwZYFwQUdWVKWk0BbvkDi8Cm1VQsgg51CXnapxopXk9q27uGJEuxGRpgk6itBRHDRNSwtS02y2aTabRI2EpmzSa7fRscZlNuhqCjHdgHg/JfvvZac6gbOeojSUxgSzLOexxiFUuCeCaLuv7gVZjd+KVTvV+p1tc2tx9qCd7KcVZCllMN/Yj79QSO+RxuLzgvFgwNr6KiKKsACVSU5eWA4ePYISkrPWsbGxwfDKVR45dpILFy7w+Pnz7Ny9z+al6yQHe7SWV0AIJlkWEp0ix1YSDPU1y4oC7z1ZlgWw34cqvgW2H2zRXO79yJbiuRkXBEG2Q1AlWKGQJ2vQ7Gdz7f+pifnuBgAlBGsrC5w4foBL1zJknASddCHDPCCC0WFZOdnGcTCSajabQVtsOGR5cYnheEQUBWA0TRM2Nu9PWR/WOlZXV7l//x4AaZriM0dZGZFYb9EiwubB8fnkkQMsLydMJhNU1GGSG0rvSebaPqcbwZoBPx0XNXATfm+t4Quf/xy/95WvkcYRjUaDtNUhzw2+GtP37t9habnLnbt3iXQUXH4Ba4KoabudIqWg1WqFPUerRVEa0qTJeDwhSVLCFiqAvo1GgwcPdvHeAIY4aVDkhqbyCPIKBxYVkPfhS8B+GmJWVJTV3sHjvK2MbiTCObL8Nrff+Q4rSye59M53eOIjxzm6fIhPf+w4mbrP0B3l//PP/yWbk5wzp44iSkWzEXH//j2OHjrFiTXFP/j1L7LQa5E0u4i4hysP4fWAi9/7On6c8Mwnv4hsJ8QoKKAsJjTSBs2GoJzcQjZiPv3zn2FnZ4O/8w/+j+hoiYMnzjEZXOXbv/+PyYsdvv79LW7d3uaRZ7c5ebqNlwrlCy68/m2Uk5w8/hgX3nyZxx8/xpW33+T4qVNcvXIBpQy95iIezyTL2NrcYTIYY7MCnGAwLrl+f5dma4WFxSMYk3Hh6haJkCwsOko/BrnIoGjQn9zlicefRTq4e/8mvVYTrSvdSwSgqi1CVVDdn8d/YuEFOGu5cfcml7duEy80mAzvsrVzi97qASwpm9sbHFk+iBZiuuc21uKKcJ80GilKad7rQvpqXz/JciaTCZM8YzSZkBcFjhlQVpsLR5GqWIQFg3LAc997jo+cfIJu2kZWOU+Y1cXPcv7/oY2HgdT6sbWWr3/j61y48CZaKawJgGUURVVxdY7sVLMfK5Zo/b7z7fz1c6fGuBXLvkZUwvuF3EvrCK01o9GIq1evYowhTZuVJ8yMIFIbTdVRA7fe+7AvfujzP2wxr2VcnyVXg3oCcBZdFnznK3/M//Rf/z8ZX75M13rGzuB1jJ1Y/vg3/jkTb7FRkJXzPnQCRR6aUqHLnKYSRHGYb7Kqk8mXlmyU46TGNhs4Efx7bFlSTkruuZL7aDLpaQnJ+nKb9HgP0fMMHjxgtHkbWZZESuK0pB8bhiYnGpZ89HNP88I3f8CDVy/S/XSXst3BixgB6OGAW3/yTcT1m9zYHHLm138d127T8oaxtZhII9DESmCxEGm8BTVxGAl4T5S0WdAp71x8hzhKOHzkCF4KvBYY5wIpqv4nxJQ7ALN7yO9ho863+T8kCTr3HvOv/1mO98dMrRaO2hm8bgOfamXOMRRrarsnaC++1yKzR3OkbmGv/ueqSaduxVYqOJzPdFPD+7fbbZy17PidMFFVoOPy6goH1tc59+hjDIZDRuMJN27cZGPzAUnawJqCfn/I8mIH5S1UzLbV1RWefeZpXnrldbJ8ws6tPqW19BZ6rK6uEkcRu9vbjEcDur0uO9vbGFOysLA4BVF3d3cRQtBqd4kiTaOR4r0jSWMKYxlJTafdYbC9WbUi12zQWVVqdsLC8dQmXbOJUVbthD+8yjtlDsqHXxveb75qVTP75pmtNdhkq3aEGvyyNrgJCSGC5IKffQ8pglC2ELLGHz7wiGJNs5lWCakmTSOKIkdKUY1XXQH2FikmlIVhIgrw87c8FXNyngVcA/pV4lxdE+FrHZ2QQCspaEQxwpfTqqQtC0rrMMYhifBWgTCU+OAOSHAGFCikBKEkCMnx4yd56ZWXuXrtOkop+oMLgCdtNPm5z3+RVrfF4sICQii67TYaw0qry9HVJSIJJp9gyhJnCrwz1T9XAcVgfXXYSlTs3GJ6BuqNZWCHgsQShFcdxhqcTyiNR2mBKR1ZVhA1EtqdLr1eBxkJEtng+NGj9DpNdrJd3LvGMdXmNQARezYezmKcxbqq9Z/ZVCxEkJpQQgTQVwaWtHxoc2RdfYC13ESo3JrSIGUA4GQlU1LrN+/H+wsJ+LygGA6xZY5vpVhZVReNx5clUkKiI8rI0330BN1HjqG8Jykc4o03aSx0GJQTso1t1noLZNaRFTnjLKfIC0pTYqyt5hIZtHq9I45jtI5C8uXCJjRSknKSzSXJ/+HhBRjpggTHKKN/f5OdzU2WV1ZIum2SXg8rw+ft25f9aEN4WOi0+PjHnuDqze9gSgNI8B6tIqwxtNtNxoOgIzYRGe1WmyzLaLdaDAZ9lpaW6TSbaKlwttZpF5WZU0G73WJ3d5fV1QOh48N7Wq02u7t9ojjCYPG2JEoiFJLClNy+cQ0NHDzRY5gXDIucRlNXYzTskVxVYJvxUevmpdDqH0B5walTJzh96hgXrjzAOkuz3SZuNLCVDEESx+zsbtNoJIzHY4wxwU3YGpKkAS4k+mmaUpQljWaDfKdPs9Gk1eowGAyn63bQnjehECCD1vBw2KfTbmIiFbSwZYR3NWNvP34yUa+NOuypK1MziUBZSza8w+2rz3GwuUCUNPnSf/R54sjS9ApNyfZWzh9942Ue7I6QWc7GrTvkoz5HDh2i1VrkqUcOs9Aac6SziDMJO+MtGrpEtjqcPP00Seb53te+y2Zf8vm/9vOkqcTLglIZTKlpmne48fIDorVneOWV2xw7fRoRNxF+gmi32L2bceiRT/D9b3yLS9c2+PW//59y7tRJJplC9RzX3vhTnM04d/6TlPmEE4cP8uJL3+bEuZNcvXKNxc4aF177Lo+e+yhKJmztPODSlRvs7maYwiGjCItkdzzm3vYQ51rgY/Bt4rRLGS8xKcaMxwVbuztYZ3ni3DHeeP1lVCwQyrG1s1lJZewlaOzHBxvOzYyHw9basbm9CVKykrS4s3WTUoNuNrl26zZPrp7j3NkzaCEQPhgR1yI7Zi5PaTRStIqYv6bOQ1kaRpMJo/GY8WTCbn9IfzAM2qg1JcLPGFRJFIOAscmZuAnPv/Qtnjn1FF/+7C+RCl19532Jn5/VmDFNZyCOc45XXnmFH/zgB8HQT+sAJvkZUSh0hIT3qEFRIWTwHJkjJT0MDAXiVCD6QDAmrg2gtY6m+fjW1hZXr17Fe0+apsSxRoqglVq39c8zXevH9ffr9Xp7yE0f1qhbyoPxlMcCFov0nsg6nvuff5vf+i//H7Q3duhFsKEzEiMwkyGTSKAcLAuJtRGuanE3wqGUINYQa40yBuENrvRIC07JqpVe4aVn6CZMTDmVt5vklo3Ssi0VeaRIF1roXko7dbjNu4x2NtjxY6x0LJSabSnYanjKkeFzp5+ht9hh6fQhNr7/OguPnmaUCh4kXVoyplFKmvdzOn3H6K3rmPsPiLstyoanNCHHlblFlg4tPCKJsCrCCImzRehitQYlBJHQvP7Ciwzvb9HpdijLkrTdQiURS71FEhUhRBi/s0JSPY/D7IfqWjx0bUTFbq2JVQEvnP71xzwyfnzx/jRThQgMzunkMWurDdWW2cQye8letsnDv9/zsw9giZAC52ZgHlC1cvjK2TssmlEUs7u7S9JoEDcaFHlBt9fjzOnTnH/iPEvLy1y4+DZps8mhQ4eYTCZYPEWek2djBqMxK8sLgeUmQ0U6iWMeffQsb1x4i93BkKJaZPMsI4k0RRTapsfjIWkaMR4FwLfX61KWBdvbOwyHQ7q9HlJIiqxACM9g0EdHCUnSIm00WFhapshHxHFwE93LGGUK7NTHP99GP1sAag22dy8KNYg3P6E/LMEQnicriYG9zFfn/VRjctbiL6ablfpdRL2psYEpW68wdaXWu72A5AcRcRzhGoFxq7WmUZY420Cp0ErhHZRlSZYF986RzgII50OrZWBLzswJ9gJ/8wxiNwX1kMHkSAiPEgKFx9vg0ihERJHnGFdibR4mLS/BKaQMgs5SOIQXOBRC2tAVKoIu8W/+1u/wkWef5fzjT3H06FH+9E//lEfPP8ajjzzCyuoSCMtolLGytMLZk8c5trzI2kIHTElhHc6YGZDqg4ar9UFb0iKgAr7DeDF4ataXBiSldXgHthzT67ZwRgcg0oEpSrROMMaST3KEimi2U9JGDN4hlOJjz36U/yJp8S/+1W/yzjvXEGXd4r+3ehXYzXZ63q33FZhq8cwqg6Kq8ikJvmave09UXV9gVgCgdpiv56uAvNamatIGcEaJoCe4H+8/PJ48zzBlQavZxGPx1eIY6wjjAiA+siVtmYB1IAWRlwglOf7oWbySNBd6rCwu0t8dEC0ssLG5SX84wFetKXWyVI+ZtNEkrfSgxtkEKaDRSFk5cABJuN6I+ZaR/9Cj9NiiYLL5gAeXrzF8sM3w5j2Wjh6C7haNlSUWFhf5Wd4A/HRGaI/6uS98gt/7ygts94M2Xhq3yIswRw36fRpJizzPUUoTxzFJEhiscRyT5xm9Xo/hcEiW5SRJgzRNGY1G1XwXiojD4TC00dmwP3DOURiDq02ovKPICgoT0Wq2iLyrpGQMg2zMsuygLBWgXq+lc4XRal/pq4KblArrPd1um5/7uc9y4/4fB8a+Uly5cpWlXo/RcEg9F5ZlQaQ0XrogyeI9rWZK2miQ5znOhnV6Y2MTpTRZlmOMp8gLtIqIdETcjLHW0mymWFuA8DgRimHddpfhYAgHEjwR3ovg3PrhJLX8RGNmMiUQ1T5EEEz6xv3rXH3jT4nMfdqNFiYV9PQCpRNkckjLC2TcJneXOH3yCPdvPODQ+nEOHlrixtVrfOSZ8zTiMd6MKa0kjiJajS6lGWN3Mto65dyTH2P18Bm+8dy3+dq/+WO++EufJ+0kyFIhS0FpEtq2RA3ucvzo4ywfPIbyA6TvUsqEtLPEpXybL/7tn+eXfv1vsbja4uZbOb3mGrdvvsTW/Ws88ewv47VCuxyM5rGzT3Lp+gWOnzjDKz94lURbXvj21zj32NO88urrXHznOuMi7FuaUSU75Q2D4RhhPFIlKGcYZLs0XYQpc/Jxn0mWcf7R82xt3wNRBIknHU39AvYO8HrN2J/Hf/wxI2wURdiHS8CKEtPQNLKEKOnSO3KGb914mQe7Yx47cJafO/tpulEbbx3OeoRUVW4YNNW98GTjDJyn0VJVy3/IH/O8YDSZMM4yxqMxu/0+/X4/zJ9hYzklieBAK00cxewOd3n71mXKyPIg7/P8K9/jiXNPcWLlCAKLFEHGy3+IQauf1XiYuem959q16zz//PMVcUxWDFCm41VNAdC97fXzRJuaqFRrmdZYSW3iXX9W3Yo+zU295+7du1y5coUoCt0qURSYqkrPOipnUgHhNVEUTVmvcRzT6XQ+1CBqHbVWqvKgHDgZ8CXtPPffucLv/tN/hrq3wZKRFLYkdxkN2aQlYkrtca4k8oa20UggQmPKAiscOq0MuJ2gzD2pUjRbAc/pxSmRsaAcVjh8FDMuC4bOMGxLVCZIMkvLRzx24AiHugu0+wW2PyTb3Q4s+dDwCx7kpGCt0+WpZ56kmcRs3bwNVpJfus7q8gJW5kwSTWNpgUe++AXe/JM/QhxYJF7oUghBRBTyMw9W+cCgFSBMjnSOSMegFJkp8UpgROiJKY3l8uXLtLtdMmOQWlG6ktXFFZa6Cxw9dIRWp4MjdOyFjlcfPF9qYLUmnU3X2llmNiWw7ilozI3bvbDYz0S87zZ/Mf/znAhzDSrNs1LnX/de71X9NH08TwB23k21QB4G47zzWBzel3jvWV9Z5rHHHsUYyxPnzxNFmiRNieKYn/vSF0EIbt68ySTPiOKIPM/pdrpsb+1w7PA6Skc4Z6ZgcFHkjEdDyrIM8gNSkWcTtrYekOdl1VYcErF+f4cH2w+4e+8uS0tLpI0mHsiqNlWtJc4EzcpJNsFXYuZJHOjT7XZ7jplaOYzP0fPDeZJUp/Xd52Ku+jQPqM7HjH368MiccyGcA2qtteR5zsgzTU7nP2/acgv4h67Rw9ICP4k2A6VUqOCpiFjHwcQGh9aKSGmMsWSTDIlkkpVoWSctEMw+KsC4zoHf4+vvHdOztk4hBJHWKCTWlmEMeYctc6yxeJ8xKrYozJhYarSP8D44/gk8hko3VQbNYC8kcdpgZ2fAvXubLC+v0Wx2abd6tBstbF4gYkmSNFhaXOYjTzzBgWZMO02wRYGzLhyTMzgX2qWtd1hXsVKr+zZsSA3W2aDJ5x1CKBppq2LLOpwpaKYJQgQndedhNMlAKhqNFJzDiwndTmPavuIQLPYW+fzHP05mC/7xf/3f4EqDeQ+geno2q/NYt/WXxmLmRd2lREmJlLN2/vnNy/zYmx9/4W/V76mKAIQCQVS1yOwHexnaQjzEMPczYEjA1nCXhThBq9CSaoxDSYUUgrTRDGCoKciFR3mBlhorPJSWg4+cYiId7V4LOcmIXcF4PObKlSuUZYmWilqDpy7KeDxLy0torWl32uRKMniwy2Aw4NCRI0wmE4oiJ0ma9bcFpkXTueOaW8irx++FHEkPncKztbnDxu07FM6wNepz4uRJGnHMH/zuHxClDf7e/+Z/TblPTf2h8V73+Z93vwkfOmBWVnp89rMf5Q++8l3y0mCKjCRtEccaj0NJSXexS3+3T5ZlxHFMv99neXl5yuZstVpQsUKLIrCdk7gGFxvs9ndDiz/BgVcgKIuCRquB84bSlajIM5oMsWXJZDCgsbLObn9EfzLG0QlNwtMK/ZyM/vTQq+OVQSst/Cj58pd/gf/+X/wB7YUm4/GIZrM1bSMsy5JxPg5j384Kq3EUBf2tPKfd7oRESmm8h9FoTBKn4W+tbjBnK0rSRtCZNmUw3hJSEUWKdrvBm29e5NyRc8EkAYFD8yN1ctuP9x81qYBagxdwimy8zaULf0rD3yI2hkL1UWNHJFsUhUFGkBnL8EHGUhzzoPSsrLX4+S9+ioXlVX7xF36Z8dYVGpHFTBKEEmR2E63bRGaRsdvERTmZVahOwmc+/xFe/O6LfPNPv8Fnvvg5kmaHRHpKWTKR4PvXSYRGmEMI0UUQum0WV85x/vzf5t6dV/H5Hba2U44/8osM717hzq2rPPWpv0rqU0o9wZYW5wRx1GKpe5TtB9ucPHWS65eustCJ+de/8y959eJtRhODQRKnEqkdWjq0EkxGGSIDn0xIpEDFFjuxRC4lIUInko9/7FO8+eYPuHr9NkcOH2UyHtNpaLa3d2m1lwBXba4r5vtfwI93P95n1K2gcxtq5wJhYzzOMKUhTiJubd/hXjng8RNn6UUNMpHzxNkn6d7u8enTz7JY9sB4jHOVDlWQL6lLDs45hIdskuNQtFpNrLWVLmoWdFKzjNF4XBXYsup7Vca7PrAEPUHyJdKam7ducG/3ATmWCSUbwx2+9q1vsPILv0KrGdprQ8trZWBb37s/NO+ZJRb1CriH2ypqssr+fvRHGt5PTUPrmMczADY3N/nqV7/Kzs4OQhKknOrrIWo9SIV3bvpiAVOQ1AN+rt0f/NzPobsxivbqnaqK9FOWJZcvX+bmzRskSYxSyRRI1bryXqlCSjl931qGL4qiStu9hdZ6mhP9ZWid/veJmhKFAOEqdqoPxUrpHC987etsXLvGYSQjU1AoQepSdgrLSqtHqxiTRw5FiY9KsJ5YBlmvkTUMc9DdDmmzg0DRaCfoqCCyHoYZuqjNeQXaQ6YFnbTFopcMU8O9MkcKzfGDqyytL9O/eYvdzW2EcTS8w0rFREGeF6zKBk8++QxeCb759W+weekqzUfPYK7f5frmNnJpmROf/ww7i01Wf+kTnHxslYkwTNY65EKiC0XkPE4YZCwpnEcKjctLPCWlMyQiItYSIwLo7LxHxhG+rLCAVkrpDMrBzm6fwXafy29fYXl1hUNHD7FyYJm02pd6rwJLe3ol5uY4P9dhWl0riQgdqJUB+iwHlTNcsPrPnpE8feDf65c/kXhfuwdJMD+anoC5Nn94mND73jfwuzVKwPtQUXG2+p0LJ1cIDdbi3QQjPLFIsKXG+oI0MSytHGL90CHOP/0UR44cpd1uo5Ric2OT27dv0+/3ybKc23duU5qSOE45cewkq8vLeKEZjScMxhOWeykChVCy+qdq9cywwGpBUWb0BztY69nafsB4MkRFgkazyWgyYJKN8N6zvCyJo5Q0bqFEYMqknZhmL8GWJZ6Y0lqyLCNRLjATpQS3F1Ctz9UPY/TaaaoWQKHZ+a6G71zVrG4DCBuP+pwDuAo0Ywqi4oOOXJ7nmKIEIacVMUHF+vIe5+cd0Ks3FPVn+7kF5icDpkY63NSlBu8CUCZ1YFQaGzZLk7wIlT1qbTKqfY5A+EoD1tfFoariyGxhnYaQSC8QzlOqmHakSP0EkxuM1wgzwpYGg0bLiOzBNuPxCKIeqbc45cPfvMWJsLhWyD5JmnL06BFGown3793FlSXdbpfjhw4F1rBUkAfwt5dqThxcDzorQkCZVeC3nVYznfOV4RMgRJDgQOC9xbkSY8rAODVh5VGRJdKgZNBx1TKYcgFYaxhnIyb5hE63jRcaM8kY9IesLDUDcF3kREqjYvj4009w5OA671y+gTRUmlMOj8MJh8MF+RCvwqbWysCuKmyQGnA+sA/qljBqJljYTBvnghA2gdFqSoOAwC6vxqSWoQjkPAFoVtWmRARR8v0IoV0Qag/CCKLaT/pKR9ehnCMfDhF5SYkiEwplQccRriywUuGsDNeUcF3QOjC4Szvd7KVO4pwhNyWlK4mEoqEVUgXZiFqOJKyv4e5b6PaI4gRnHCYvKF1Jt9uEMqfRXkZXrONZ8vbexyiqecvvmWL3LspeQInjzVdew2c592/cZLHdYjgccPHGVW5cv87y8ipbDzbprK78aC/Chzy8EHinSYTnb/zi07zw3W9z43ZE5iVFuUvXLBA7CUlgYJrSsLO9jVYxWsf0+wOUUty7t0Gr1UYIxWQyIc9zFhcWQpKdZbRabbQOGtB2NCJJGhRFThwrlIzQQjHKSvpFSSPRPH7uLJFW7AwM9/L7DHZylJMgDLauzhPA4Dl0tTomphtFAXhn6TVafOnnv8CffPttlhsrLHVTth7cZjwa4vB4Cpz15OMSHWmiSKMjjYwU3V4HoSIMjkgJdBQRRTFWwOLiIkVZMs5HNNIEYwriOKLZTBnngVmbj3KUlDQ6glazDT7GyxJh5RTQ248POMIkSyE9GgnOU/Z3eOuV3yPffZU713e4dOctdu5n3H0wZGhhMnSspBGPne1CEnHzgcW4lOWlZbJ8l2MnPsLbbz/g4ksX+PlPn0BKTek8QmmMcyif0Wgvk5GTuV1arQaRjDj75Am+//z3ePHrf8pHf+7zRHoJdAOLxVmQ4+ts3PwWaye/hFeCSDg8KasHn2B57QRZlhNHcPfaS9zf2OHRT/4VVJIgbYYucmzhsd5gyegttRiOt4OZXCJ57Y03OHH0GC+/egW8R0cJSoCUJY1GKxSBvWFSjtHeoeMWadogkRGTbIyXMR//1Md4sHmdS2+/zdJCj0h4VNpBCsNg6wYcOYEX9S5Cs2/J/mOKCkw1eBQOnKMoHNl4xI7p00gTWqQcXTlEIx8QCUnUVAz6E/y25RPHP0HLNNBGUkgTdhXeY8NIDF1LVYcSVZu+zUsKN8TYgjybUOYF2SSwU/uTMRNTYJVAeoFyntgFeZaJDgbHJxcWcX7Ev37zq2yyhXIRpbfsmB2+/vo3Of/Uozx98vFgY/ZQUv9wzlMTAUSVfAX9Yw8u5H/1HicwtMJrPoT414816q2gkYGlKAgSYdIHY6JhNuK5r32VjY3beErwAQsQyOk1UTIYYVbk5SrHVYjKCHKeoSqlwDlbseBnj4PceShsai2RyjOZjLh48S0ePNiqwNN4qgMfDKoDeFsTheYJJPP6rAALCwsfer1UYGovGHodQ5avnUCVGqck2/mAOFLYrGDcjCiFo2VT+qJgw5UcpIHKPDbVGAeZloxaTToH11k/fJgjJ09z4PARur1FvFCI2GLtLn6zz85bVxjeusPwwSbGGsxggMxyWkIRD/osypRud4HNIuOdN1/DyfOM+gV9J0nTmNa4YNt5drVnOBjxyJlznP+5j3Njss1wa5NGYRjcvsNWFHP11Vt0equcOXWGornGRjtm7cRZEiQDAKEwXoLUGGtJVYTwwQsnEx4hNLbwEAfWtbLhXBkNZayxAnKlaSmBdCURDhoNjJN4bbm9fYdbD26w2FnhsUfOsbLSRWrwSlCaCs+oyX6Ei+KcR6pg9OUrqZa6wBAY2xXDVYoZ1iVgb7L20zm2378BVV3BE3LKRJ2nsNd6M4EF8ucwT+bAwvqpdau4J7RtRKFvE+cFxlkQCY1mg3/4d38NoVusHz3Kg51trl69ynAwZDAYMOgP2N3ZYZJn5EWB1prl5WU67QQpAgCcGUdhLJtbWxxYOopEYqrW0FDhCWmEVIo0TnDGsLuzHUykBgMipRDWgzEI65EaTJlz794dtE7Y3t1msd2ht7jIUnSAtfU1xqMJk4kliiVSTfAmn7IZ5Ryj94dVkvboSlTAgpiyQOvnvPs8z1rzZ9dHiBpofehaCIExJVk2wXuJsT5oUdbsVhxhD1Abwjy8efBVG0M4l1L8+ePgRx1KKWIdAQ4tAR/kDJTSOATS2AogFlMG9DyDzdfno7qBa6atF7Nq8Z6igCCAqSKAoFIIpAggZiWNF+QxvMCVlo07dzjc3yFtt3EisDcRKpwxISpNV5BC0W61OHHsOJ1ul/OPn8cZw+LiIo00wXmLcTDYHVCaAo+jkSYIEfSjwg52JsHgnJsjG4X71zM7B1Mw1YJHI4XEO4/UkjhSRFoFrozzFehuySYThuMxhw8dCnqWQvKgP2Rx0Gah12K3P6LTbBBHinaU8umPfowrl28glAyi7qK+PqICH6pSTPU9rXEYYynLIDkRjKgczgu0kHuvQ814sOF4avO0qdbzdJ5SWGMQcqaxaqz5wI3SflrDCyiqU1GXZ7wggKsiAKnZcMDdK9dYaHWJdYTQkswZEiFRcQBKFR5hHEZW1ilCUBoTKpYiOOpGURTm+SgmiZtEUcwvfPEX2Nkacuf2PYqiIM8KrDEYk2NseL3UCqU148GA4XjMQqeL9Z7SWJy1KBkBM1kIv4ddWx0n7/HLedZCxb41CO7eu0v//gaxFyjnKCPJg80HaKWI0yQwZfbjRxxi+v9HD69z5vQx7ty/jfKS0jvyvMABhdmm1W5XppUa7wQqCoXMLMuqFjpJHCfkec7CwgLj8Zhut4v3np3dXbRSLCwsMBgOaDRaWFNiypJmu8todxsqloOxjnfefgepBM32AZz39PtD5gttcwV4xA+FI2uNbFBa0ekukCQNdKzZfvCAyWiMd5bhZIjSHm/DN5BSkCZBlzpJUxqNFl4odvsD3DAjTdsA2ElGHIWOks7iInme0Ww2KLKMKNI0VZOtrQ3KfEI2GdHtdoI8TpX0TY019hP6Dz6kw1ISYfEuZmhz7l74TXYvfpU3bxVceeMK12+/yfUdx817MBxLjqxEPHuuiU+W2egbRmVMo9Wi1etRmJJXL7zKcNjhey++wuHeiNMnHwuFcglCRBTaoHyGzASLchHlBEMrWGl7jp04xltXbhJ9+wU++snPIRotYt+nLBOyiWVy/03i5gpLh54O+wVV4IVBiA6NRpvXXvkKxWCTJz/+i/gkIRYKrATjcS6vOr0UzhYsLne49NY7bD3Ypddb5Otf/zqPnjvJ7edfRcoGQjjiKCWJFFJ5XClQqkALRaOZ0G42EFKiZCiwtTst3nz9VUxZkCRd4iiitAqPo7+9tYcEUBew9+NHHx6PcJ5IBlPX3Fn6ZUY/6/POg2t0VtuMRJPItljpdtFCcr+/zd2N+xw/dIzURygvyLFYH97DeYeddlrZKXMfV+VJtmQ0GTLJQ7dKXpSMs4w8LzDeBYduLRBWoJDV3tMhlaXb7bKwvsxzb36L71z4fjBecQIvDVduX+VAvMQ7l97hmePng0+Ehyrt2TNlGmMoiiK0h2uFlAolgwYnNYXDe6wPUlPivTYq+/EjCjFtq9/LTXWYouTFH7zIpXfewVXkLghmaHUL/vT11RyhKjKZrwCeeQOq2vy21jGtPV6kjKbvVxvojkdj3njjDfr9/lSaLmAQtQaqotFoEUXx7HvNdbHWRJn6sxYWFvZ05X0YWalAVdgGRdAdd4Kwj5ISpwRbmw9Q3mNiT7uRoEZjYuVoSc/IZYydDoBfK6F56gQnzz/G0cfO0Vw/gGq3kEkSupaFxAlwyiPFMulpWD1/loOFId/ZZXD3LlvvXGV89SbDjR2iQYF20DSWVW9xuefuD15H+5ROr4NPG2RlH+cdu9mQ9tISxz76BJmCqy+/zs6DbXSjwUof7t+5jtP2/8/efz5pluX3ndjnmOsem7Yys3xVd1W1t9M93sEMwAUBECCl2CW50kob2gj9B3olKUKvFAqJilDsC4VWopE2glyaJXdhCBBmML5nuqe9q+4ub9M//ppj9OLc+2RWTQ84IAY9S0z9JnqqKs2Tee9z7jHf39cw3LrDR99+iYdXfonSgbPBok+mAbAHG7x8XTh3BfJcfeZt8A1rMd4TSR2YvFox08FWSgtB4ix3blzHmooj6ydod5YYtyJ8yyHKkvFkwg9ef53+QsrxjaMsLSzRardxIqjB8iKvLfkgSWJaWQZSHFju1Af/BpcI2EttL1l/HOEPznICPh5Q/dluXH9yMLX+U8g6t/aQpPswuPrjjhCHuyXN6zWeH1nWZjqbBjagkiidcHy5x9mzJ/mz194mso5WawUoeP6px7m1O+XVV19lZ3dnDmSNx2NG43GYXKTi5PETnDl7hvWNDWZ5SRpnRElENDMIGj8UAUoS6eAd2QAuSmqUitBSUeQFw/0hRVHijcFbz3hviMljyqpCJzFJFOOoEFJQlI7RMKcqJwjpGe8NuHL5GipKWVk/Gox+a0/HGr2bA6v33+8f9exsxCTUwOX93k/33u8DkLt5zYPB1kzGDePP2YOQpaKWQx4Gyg8HBIWAn8Bda36vAPAyB2t/FoNaK41SMsgnmg6eUAgRGMfNg+th7sE7f0IP3cb7vWabpsj9AF7AqWXofNWfq6whQqCUwBS+ZkIasBZrClwxDaFQdeqoCH1qQOJcAALKqkQrxdrqCr/6q3+Dix9c5ObNmxxZX5szj8uioChyklZKaSqk1mFSrL1u8fcGiDUHhqDe9jhvCL9WAKGctVSVC4Erskm59QF4lHI+boWQdTBckKAYYzBaYfMKBOzsDINENZ9Rzkq67QSpFAv9LnGkyMvgv+Kcn0u4GxYiwuNk2CyUtqKsSqaznLKqqKoS51K8PwA+lVJzZoK1lqoK9gpNSRk2qVI03dqD8au0nsu6/I95hn4ey6gAbqtazSQ8aGtR1sJ0xv6lG9x4+z04eZJ2v0vUzhBpxMSW6CR007WUyEQgfGCKW+fwziGVmIcJGhMOGs46WlkXYxwL/WVefOHT/M7v/B5RJBEijC8dx9iqJMsyjIBxnuO8I80yVBxROcfe7j5Hi4pYp2EukuJQh+neueiw1F80H7ivhA+z62Q6wzvYHw3x3Q7lUFLOcvq9Hitrq7Vn6oP6aVYIr2wmXclTTzzGH//ZR0RZnzhqEVGHQllDVVXBK7WoIBb0FvqUZRgrYePosLak2+2S5zlplrG3tzc/uAAURTFvwvR7fYajEcZY4iRllnt0FCOVIG11kNLT6/fZ3N/F2mY9PTzODmSAzfhrtA1weM0UvP3e23zzm9+jMm3Gk1GwAUoyKpODK7EzUEmMkB4dKZSOWFk5wng8RcoYpSOWlxLyWsLqvSdJMqQQFEXBeDxmZXkJUwW7A2sN+TRYEAlhiVRKqjVaRfXceODF/qA+2QoNaxmUGYDFsX/l69z94Fu89d5NFlfPIOyE3bFkeywpjKAVaVYW4OzZDYxbZGuwj+y0IY5YObrCsdMP47zj0Scv8MLz/w9+9x/+H2nxJsfOPI2LJC6vSFVImFaxpihyRqMZncUFCud46ORj3Lk94OqdbRYvvsvxR46hkyW8DPsChGL78su0+yfQrVWktDX/UDEdXcLmdzn/5JfR0WrwUqsKvB9hbVkHEBEsiKzDVrDQX2Drzm1mRcFjjz/OrVvbvPDMo/zgtbdJsi4qUqRJQqIh7SuSJEYLQaQ9Ioqh8mT9Pp/78ucYbe9w4/otllYWyOKI6awgybo4ZxkM9uakhJ/1Aeyvexnh0MJigWllEZXBVENExxFXCd99/WXOHTvJZ059ilhJtnd2uLW3zcnjJ2g5jTLgpMNIEcg1zoaQPmuxPvx9ruqrp2HpLbPZjP3hgFmRB4WSDIx7LSWRh0qA8gKJphLBB3Mxa3Pm9Eku7t3gH/3BP2dzvI1wGiEsCId1BdZWXLt0Ff1VFcgAjZjs0BbSe894MmE6ndahzWH/GUUJaZLgnMOUBTJL0EkS1gkHqAfj8K+imrdGuoaNyjz4+uL77/PK938QZPeRwlZVbSXWsNbr87IKEv+GeBTs2ALmYWvgNfzdzBWfcv6+R1RVNQddhZAMBnu88847TCaTeahqw0YNzNSULKvzPurXacDY5uzuvZ8TQdrtNp1O597r/jkFVK0EbWucStSZJlpREpjIS71FskqwYBXZJEcnGqMs60azb8H2U06/+ByPfPkz9I+fJmq3qZSgUgIjQq6NqIl/Siuk82gSnBK4hQ6ld7CySu/MGZaffp7y7g77129x7Tvfpdzbhv0t2nsGJTWjaobRgnxgiWUPG7fYHw0RleSRF57m9LlzXPvgA/Zf+5CosNDu0B9IdmYjdCsoQN97+RX0yXU2Pv0MMy1RicJ5EfAqWSuVa3b0gfVdADJ1JMAaLKAlpCqixKARIWDXl8RFxeSDD9jaus3CL7RxrTZOJ6Rek2pB3hJMpGRnNGFr8w2ytEW/30dFmrKs2N3dCxkbpqLdylhZWSFJU44cOUIry2hlLdL4wFJSAObQ2D2w72u+4s87r//s1vOfCEydHxnmJER/QDEH5mxUH7zt7n+I72ehNtX4fKxvHGU0GtHr90NgQpLx3LkTLCz2ef3aFqoscU5hrQNTsbt1l93dHRb6fRb6C4CkrAxplqK0ZmVlmeXlFYoy+OW0O57JaEYcJ8wKR6vVoiz3kEoHYDX81uzvD5EihNQoFSGFJC9yhBgjpcT6EE6R4LBYlJRoBLHSREkMUgY/KmMY7u0zrRzj4YhLH7xPlKasHj1Jf2mZE8fW8TJDSBvAtLmE+Uc9Z+8FVMNNPIAqf9REu2Gf+vsG4/z7aQCl2tCg/tlpmhDHMVEUU5kqyBkOb1QOlbUW633NbjwkxT00LuCTD6ASOPChCyjn3bngaSORCFniRQjHaRajhonzcb/pvMsn7vt3zfYMFPam0xxet6wciTUgNEqHySSwVEPq+O7mTRbWN3BRkJZZLwAZfOziEJCys7NNmiYBgBKwsrzM0vIKaZYhvcNUFbN8yuLiAk4w37R5d+j3q8HiAzA93CHvHdggbbc2sFedqbDWYK1H+4MEaGuDHQQiLOLOhdc01pJmGVm7XYPSLshjvWC0OyIfT0jaCZ1Wm6zdZlpMcMoTtWJmZREgBaHqblPwjBUysJ7r6CiMtZRVRVEUlGVJZQxlVaGURNRgixAhkMpai6kqXH29BxuhUOF+QFhWa+mVs/jaXsDU/r8PSqAtgSlM3eEVIdG+GI1gljMpZzz1mecDEOUc+zt7YawYg4o0C0eW0Z0MpyRSagpTzJn+xhi0Dt6NeZ4jpGAyGqO8Z2d3j8FwxHg84KGHTyOlorKWyWTCYDCAwnD+wnkKU1LmRQB8paQsSxIVIa2DvIKWn89/gZP/Y/iBzTP9Y+6Er9mtk8kMn8/wUjCpClyh6WUtfEvSX1pEqAeGqT/tEs08VQf9nT55jDSRWCkojaXbyUKoWW3VEUURzgavu2bNq6qKsgx2EUmSUlUVURSxs7uLVgpr7dwX3HuP1hFZmhJpSVImoW0pJVEcU01yjFOIRDPY20VFY5yH8WTGPCtW1OmkHyORF4eanmGfJHn3vXf5v/2Df8CtW9A5cg6lJDNTMtjdwvkcZwqEDwftVisjiVNarTZV5RAoRqMJHoiThDhJiHSQcsVxzGQ8RkfherTSjGcjhHcIFdaaKNIsLrRQQnL10nXkZx+qWakhENF/THPhQX0C5cPc64VmuH2JW9//fb7xh2/z5Od+kRtXvsPFG7vkZcpvfOERLlw4zXS0h8l3iHXG2+9tYnodIm3pdFNOHj/OwsIqCwsZnShCdhb5O//V/45v/tP/O+LK66ycfzSkQ3uBiGOssxBlzPYGbF58k9XFdZwqyJY3+Myv/Rf86X/3DxDRlBMnnyJK+yhCqGuVD9i+/F2OXvgVrIoQNkIKy+7da7RbayS9IzgFsirAzrCuwFGFMebAW1srXhRCKrJWRlVW3L19jU67TS9SnD21ys6gRAiLcJZeq0O7o1BRjEISqRIrY5wwnD51gul4zLvvvEmv26XT7pAXOVImIMDaijyfYEyJUtmhfd8DUPWnWfMzB8Hyp5hV7O8NibTl1uAKE1Gy0Fnl8bULnD16mjSJuL23w52tLU4fP0nLRGA9Tkls0JUdsOa9B++oyip4/dfkFNGcjYSnKHPyWR6yL6XGO0ckNdIFD3epFNZWCGUBRy/ucO70cXZn2/zX/+r/xbu7lzHCIh14FWzOenGbrz7/eT59/lM4B0rfC6I21z2bzRiNR0xnM8aTKZUxRHFMu9tlqd2lkyYIJdna2abEcnLtKDEKvPq4vu6D+ilVTSAO5w48d+7e5et/+nWKaT4HJ7TWqMZiSjQM1EC40Vpja0JJY6HXBD+FvW0gyDRs1MN7kSRJ5uzVu3fv8tFHH1GWZi7db1SgURSRJOl8X6N1hI6iOWh6OD8i7FsCqNvv90OQ5iEs4edV6t8Qu3xNDpNeUEpwWoG1ZHGKqZmPsdNI4/FxQqQlR0+f4uyvfY0jjz2KSFOwEQUhlDmpQW0tZE2eCq8tpAJ0COHWktJaHA7jDXms4VhGtrzKqVPHKGd7jD+8yO43f4Db3EbEYKoJrajNeDZmlmZMppYnTl1gaWmV4fXbXPnuDzE7eywoxVQ4fEuSWYHMPb7TI1k/wod7W2SmpFSSHEEcx8jKIkUgvjVYnffgnJ2PJykF3kGkgyTfmOrLxQABAABJREFU+LB/1dZTeoNVijsX30devcHJhZTBlUus944gSdEyEJJQkOgY6QVSJxRVxe7eAFGPaakkSZqSkGJMxe27WzhnuHHjBkIIut0u/V6PxYUlNtbXSZO0BqsPSBUHeJa/751u6s+bOD+eVPPTrp/McX3ONgz/aDpt+JC8PfeFkT/u6FqXDwtdfayYMxyzLMN5TxTHrKysMp1OGY8GVMWM8XhKKgSmMrRamv39XcrZlLLI2du1lGXF0aPHSbKMTi8EMnx46TIWH3zROh2SNCPSCXv7e3gv2B+OaGkzZ9JKAri0s7uL0hohoyCZNhWVMYgy+IsZ52qPP0PqYxSE0Aqp0JEOUmTvUVoyKyu2trYop2Mmkx3IJeNyTH+wymK3dU+wSgMM3Q+M3s+QbG7ix7Gcm+8PLxEG3VymfghMDSzARqrv5uBTkiZ0Ou0a5JiFz/sgqZlrxGuyl7QCWSfDi1rOHmT/tcxffEw62ydRgV4zB/sPg/j1Ixm4Q3O25gFgfHgT8yP3XtzLEj7MwnZC1B6ggZUxqxxJVSFjhY4irJ8iCZR2DezfvYU3BdZlYQwASZwQpwlVVbG3t8dwMKAqEj713PMM9gckSYrUEUprsCUCyWpnBS9gPJ2SZRmz2ZTS5cE/ioPf/x4vXnlwYHAusLa8bYKn/Nw4Oviruns8V5VumKiBXRXVv0/D8s1nJVWRE8d9jAtdrfWNLHRbI8GXvvA5jmxs8I//4f+PK1duzOcB4SUCh5AHvGtX/w6VMRhryfOSsqwoywqtFDiBrjc5jeev9z58rnnvGimOdXjr69RW8M6j6oXjzw8L+Pkr4QNbQzTSZeFwwmO8Ja8KIg2rp4+i0hiEQnvBWrfDZG/AaHePve1djDF0Fvp4AWmnHdik1uK8QynJdFwxGOwzGY8xleH27Vss9Tvs7u2zsLTE6dPHOHn6GHGccPX6dXZ3dzFmlUwmLCz1uHnnDpP9EcJCkmW00ozSVfiy4sYHH3Hq6SdIalaikI1fz/3z0I9hrDY3oW5XSSHpdjpcv3UbpSAmNOv6iytYJTj36CM/2iF8UH/58k0aaPhnt5OSJpJCCMoqWHMkOsa7sA8JQYmS6WxKnKVEUTSfG5IkJk4SZtMp1lq0UvM50RhDp9tFSUmn3UZFEdZUgc2uJEVpiaKYbq8PZooxPjCrdZjbr9+4HTZ93h4iQXvwYi46adQjcwa+gytXrvKP/tE/4crlawzzLgtrAldVzKZDinJIUUwRQtGKNUmcEqcpaZKhdUyRl2RZi8l0SpqlDEdD+mqB6WRCq9VC1OM9iWJMzegPv5dgMpnhELTTDCtyvvONP+PJxy7QabeRMgAS0luciB7Mi59A3X+PBSbIBr2jvP4Kr//wTc49+WlacsA7H93kjesVv/DUGn/3N56g1Y3J0jOk8QK3Nofsm28zRGCkZmP1CMv9RbIoxqMQpUbqffTCOp/7rf8t3/6X/yfKN0c89MIvUtkWypcQJ1inWFk7yaV3LnPn8hucfvbT/Np/9b+inFqyxaNs7ZR00g9orZ5FRS2qYsJkuIdkSrawTnfjWSIZsbd1icsfXuErv/x3qLTEUQEzvN/HOY33Fd4VeGfxTuNdRWVmGGuZ5Z6iNEzzkm5LMNzZ5uHTx9G37jIaGbrtjF5bk6QJQoNGYnVGVlmipR5H1o9w5YP3ubWzy6mNYyRSsD2esdTv1GuCZzYbU1UFKsmoDyUf2wR5UH+xun88e+9RxlMUFbPJBC0dM2G4srVDu9OmheGZExfopC129ne5vnWHM8dP03Ia7QVOK5wggFuEBryx1GSIsO8zdYPfy4NDt/TBZx+lqbeEeBnOWa1ORqfdYTAcUfgJ1pUs9rqcPnGC/cEm/+Tf/hNevfkGE5HjrEEJj3XQay3yq5/+Gr/+4tdYaa/glcCKkBY+lxkKgTGG0XjEZDJjMBqyNxjgCOqGwXSG6eTI5WXSVkyn1+H199+i226x1lnmAaD/V1NNyKioF2WPxxnLW2++yWA4AAhkDnvAHD2wLaxfw4eQHOqsicPnywbUBGqQ6uA5kFLU5CyJc5abN29x/fr1uRVZw/vQWpOmaU1oikjTFKU0kY6RQoTG1yFGavM9zc/vdrtzHObnvWR9+50IHrnSA0pgvIO8ZHNnG9lrszfYo4gULa1Je32e/+pXOPrFzzLqdyl8TcDSMmzklEQrjcdjpMDJhqHcDAmLlKHJo5VCeEmkBGUsyE2FSSOSxVWqpT7LR4+xfPwCd7/3MpNXX6LPkKiaUdqSrekEmaZsnD7OxuIiF996ne27NyETxA6UsVQRRIXATh22p3n4088TP/4wxjmqskTFEldZOrXCGmtDk9bWDQIp63O3RyLwkZ6PqYqaxFBVxBKUsNy9doXxxQ84/vjDvPW979M9/RhJewkvJFaGhK8WCiE8lYQ4PghlA4+UdUMCT1yHv1ZVhbGhATEYDLh58ybTSc7y8jLnz5/nxIkT9DsdPOF+NsGth4mCvpl3+Z+GouonZqY2GF0ziThHkLn55t+uTrr7UXbloX8cbFhq8C/Pc/I8xzpHnueMx6MaQFEc31innaSYssC6ivG45NJHH7G6dpzFhQVOnTlLnKRUxqB0BFIwHIzpLy5ijGdtfYNOp8OtW3eRPkj4nXFMxhPSfk3lr1E04x3D6TRIQgi+jmWZ472jKPLATLWWuqFOnof0SCkkRRk6T3NpPAZrHGVZIbFEzmCdoxw79o3l8ofvcuHMUVpp92Pv98exU+dgn+eATekDczE0a+8//AQbgHlCfV2Hqd7AXLoQRRH9hR7OW1rtAPRZZ2vAzx8kaxOSOJ0D6w4A2WYiPwzQfuIHsjmj0s7Nu+feMtQsTB+8Py3hvznYXDcHwsscsIgaUPl+YLv5UwpZU34tXipmNniiKFFSmDAeIqlASHxVEVGhRSBWSS+Jk+B/Z70hz6f0+l1Go3201mRZRmmCcXQSJRjncc5grWE4zImTFGssy8vLxHHElcuX5s9mc10NGCoC4j2fhLwPMvsgswvMqnDPAG8Dk1bX98w1k5mce51AkArgA2A7yieU5YzeYkqmE1xRhrnBC7QQHFtZYWlxCe0l/+f/y/+VyaSksQ0RSOq9cM0RD7fUOh/eN2fn/qmVrlAyuodl3VTjewx1YJoIXUNjTQBQanDNO0AcPBtSPGAXQpgKjQihDJHzKA8Gy/7eHju3b+GnM7SApSOrZItLlN6i44jWxiqtjVWWK4NCYIrgCd0E2Dnn2L67SZ7PKIsC7z1FUbB25AhrTz3Fu++8idIB5JpOJ/QXF9FacfrUCR46ezqM19JSOsvxE0cZZHuIKmwMJoMRr37/FYo858yjF+gcXefYyRMHF/Wxq2wTpijm61ptgFwDYeHzkfOsHVllsr3FpcuXkLEmOXoMAXT7ffrLy8wH7oP66ZaYTwgkiabbjckHBuGCtMo6F1KXo4hW1sJUI1ZXVhlNxiwsLs5ZqlEUURYFSin2BwNWV1bmEviyLMH7YBE0GrK6toE1FVGkmRQFOtIkWYpVCdW0Yn11g6VeB5lksCvZ3x/jfLPfCT7TiPn2bn4ph5kir776Bv/yX/4brl29TSxTytmITAum0ynYGcPhXayXdNrLZO0WKoqIdIKzgl53gdmswOOJkxjvLTqSQPDna/zdsiwjz3PW1o6Q5zlSSRw+sP1LR6YEWRJxfGOVpYUug/19Ftp9vHc4YXiQav6zKe8FVkisn3Hl0nvIuMvRE21++PI3ubNV8tXPnON/+befp9VPMCgmzpEXFZ2lHv/5f/FrXLmxz/d/cJEzGxt0O22E0mirMdqiRIZ3FWJxjdOP/gJv/OF/QxW1eOipLyN8yAGItMM5xXOf/jSjmeTEE79Kq5Xx2rf/MZ/9ym/xjT/+F4yu3OFhlbF05GQAyfbuUtBD3HqNaRlx4vQzFKNbnD59FFQbjcUxw2HBVcymBUU+mtsweW/xzta+6JbxdMY7733IhXOP8N1v/gkvfuop3vvgAy6cf5QbN7aIhCeOFVqLAJApECQINeHRR85z5co1bt++xdrqOjoCYwusESgd1Sobx2i0z3Q2Juv0CDuMBzDqT6PuV8MZY/CVZTDc5807F4nXO2wkS3zq7HOUSJyZIaXhzuAutze3OHPiFG0CM9BJGVRMtf1KaQ1aBkAjqgN6iqoMQIEQdSiZnzfKrPMgFRIZ9rMSokhz/Mwp0laLtbJi6+ZtokjSW+nyzo13+Gd/8M9549Z7TAgHfeEMFkdLdvjCY5/nV57/GsvREsoF792gEQCQ8/PsbDZjNguZHePJlFlRhOBTBJX37BtPGsf0VQcrLbe27lCakq+98KU6O+DBSPxpl6+B7nlTE9jd3GRvf2+eASLnLNRDZJk5A5TaniqwToH5OQNqsk79tikl5+fgRpovZZD/X79+nc3N7fnrBhuAxuqwRavVot1u1z6pEUkSIxBzIgswP1c3kn9X74P6/f691/xz3AzVDtw8Vzrs55XzxF6wNdznxuZNZFmReMm+M5T9RX7pb/0mG888RZGlSKOI6/fIRYBvPI5toFn4OpjMQ9C+eYQM+JdDYJ1FOhGCjpszaaSIlICqQDpPerrP8upJ2utr3PneN+HmTfpVYNlvL3iG2Qzx7ptsX7lE6Qq6rTaUltjXFnyRRtoKU5Xs373DQ08+wiS3tNIUOfMkQuKEpfKSSOk5ztWQ6ZQMWURegNcKb1yND0hmtsJITyol8WiKHw4RWIbX7/Lo+fMQC6aRp20DOcBJ8MKFO+393EKgAVFlg68I8F6G7Aspqar6WcHTardJszbT6YzvfPe7dN54i89+5kWOH9sImEM9vzd0zRolqMHVT3Z8/bj6yXfNdaemkT8G34h6UlA6SObrNsvHUcznoGD4x/xjxgZ/G1GDlTs7O0yLnNVsg+WFBcrxkLyqapkavPbWu/zWqbOcOX2K0jo+eu89FhYXeeLJJymMYTLLqSpDnCbcun0HrRTtdodrV25QlDmTWUhCj6OoZh1KhA/A53g8RkiJqRxlWWKMAR+SIoUMDNZIgLcO6wm2A0KQC8FE1VQaEWQkSigiYCFRdPoLTKuKvRIG04qbd7cZzHKWXBuo0wH5URC6Ya66Q8zI+YalYZ3W0hZ/+HvmbNRD7FT8PWzWuVS9ft1Wq4UxJgTDVNW88+t88Bk11oTDIgSmmfMhZb32HTU2TDS2Bubkx4yBv+ryjjloCjWgZi1CqhpAdLVPrJ1fSzMnCmB+I5vXm99j5gDx/d2/w/c2anXJ+iu47QneWoT0RLVsxDQvX+ZMR2N0sgDeo+OYypSMJwOsM0RRHPxGpELqiHYSPCDH4xGTyYQiH1CWRQAMsoxWmmKNIU3SOcNailpYeghcboDLwAhvGLnhT2MCu1rI8HdEuGZtBbFTc2/LcBMFSgbwQAgZZC8WTJlTzKZQ32PhNUVRUVURcVz7VSnFs089wZe/+Hl+59/+EcgQbBXM4cOYUSjmbmbOh9TTesw3Hr9FafDaoetnuGHJC8JiIaSgKoo5A75ylspUYbPUvHbDUqttEx5UXcLjpMcIXzeSCoR3HD9xHF/m5PkMax1XL35Id3GBhdVlXBS65lEcox10oxRXVgxmIybjMf1+j431dbx3bG9vURYFD50+TZZl7O3vc+H8E5TGcOfuXTqdDrvbA/I8ZzKZEOmYYEMSsXH8OEprXKtkOhyD90z3B9y5eo2lpSXyyYSyKOaX0myc7z8sH+pLHXxMhrGG99iqZH9nl/HmLqsry6x/7nM8++wzXL5xjasfXmI/yVg8dZys026WxQf1U6wgz3L1nCzoddusHVnk5tZd4jTDViVRlDGZTFheXqYyFVJJxuMxG0c3GAwGNVt/hvegVPDe7bTbc7l/u93Ge890OiVrtZhOJ4zGY1aWF5lNJ2RZGsIEbRN+Z9FRxO2dHY4cbQOSvb0BlXFo3Yww2XC676lGpbG5tcXv/f4fEKdtVteOce3abfZ3ttnbvskst9y48QGTfJ9Of43l5VVaSYL1Aomi0+kGlr23FPkMqRVJmlLaAvD0+l2c8+zv77G4uEAcRejaBkHpIB/s93pc+ugy4+0BSx3IZxO8XQkNBBdsV5zzD4bzJ1Q/okZCooRFSUF64kVODkvuXPmIP/vu+6wudflf/NrjnDi6zhRQMqnZVBItJHYy4/TqKqu/dAQZa3QS4ZQPsnYfAxHKzRiNhgztIiY9ypvf/Ta7o4JPf/4r+MIR2ZRERkytpHX6LD4e860/+X0uPPYCr//gFX7xN/8+m3sTNj94G6FuUM2GvPSdD5m1Ff/l/+ZzXL/8JsdOnuLP/t3X+eVf/02sBOUlUhRYCia7U/7sm98haQtefP7ztf9lSVHmFKVlMp2yN9hlPDW89INXefzpF3nrnTd55ulH2doecHy9z2gwREiFlA5UhFACWVk2Tp9jb2eTfDwkd5L1NEamktE4Z3GhH1jXIqimptMhZT5lvjLUVk0PBv5frg7vixvmXaENVlpyWfFHb3yD80dO8JWjz6GdotSOvWLErd19zpw8Q9trIi9xSlFUFXmes7u3x6zIA2NPaxb6fUyeA+HM0oLaBqrCeY8BZkVBnhcoFM4apHDEUnHy6AaLaYYtHRbPsWNHiVB8/92X+P/8yT/m4vQ6+8wwFcgyEE0iFM+feYLf/MyvsBIto0USQm0qQ5yompbGvEGclwXOC4qyJC/zcC5y4ZympWRmCwbjMVk7wUvDzTs32J3s8+UXPk/iJA/6+j/9mivQ6iN6OZ7y7g9f48jyEuOTx7h+9VrAAUQILAbmYGcAhjxai7kv6r12dgckogAgCZTSKKXnrwGeq1evsr29HbIiPPPGZxQFRmqr1a4bwAf+qHEcgqcaslQURXN/1sNEmYWFBbIsu/eaf85BeU9gpQoPRoW/K+e4tXWH/b0dklkBxrJ8fJ3P/s9+m86Tj3CjpZAKOnVTw3uLEwSLEWGhBlGlEHNr/IZ/aQ64WPUxOahWnfdYH9b2SjtaQkNlqTLBJOuz/JWvsnriJK/+039Gce0q/Sj4iE8Gm4yubzLc3WIxaZEajXEahyRy4ClIIwdmxubrb7J67Did8xcojaclFbLS+FgGDMtZJCIE4NXNA+saQl6Ym7yUSK9QgJCSMpb0nSS9McTtDVl94hz55pDBa+9R0uL4bxzBt9tYodAi4AO+Tt0WCJTQ4fxen8FU3QyoqgLnDmz1nAss1UCAi2h32ugoZjqd8vVvfIP1tVWefvppjq5vhLW6JoOFLfWBzeT/FOonY6beD4zWMv+Di2u8J/2fCxN/nPQ7sFNntNttojiiKEu01uwOxmzdvYubjSkdocNtJYNxTqIlw8E+WX+Zcw8/zJG1NTY3N9FxTK/XQytNHMXofgBoN7c2GY9HtNstVJRhrKfbDVLQ0HGAIs8ZD0bYylDMCkxVhoORs3jfHOxk/WB4bMNgEgIpQFiHApJIs5SmLHVbHFnocmK5T7+Vsjcpubo/46UPLiPKEcLl9014NSh6+P6JwFbgvuOZPzyA/AGT79773Ejvm04b4A8m/WYybkBPKSWdTodWqwXO1exTV79O4595mIlah3sckkweZqj+bCqMr+YaHdTeeAnOhU65rVmKTUfSeTeXeDWw9I+Cpfeyiw5+WvicqwHVpNXlxEOPsLlzE1Gn6EVz6nDt8TibMp1MaS8FwN07x/buLkLaeaBTkmYgJEpHxHGEMRVJqoniDlVee4lIiZQHRuhKSZaXl9nc3AwA+KHfv3kfhRNzYDiwUu/1tbXOYV0A9xG1F6k/GC+mqijykjjN5h2iwPK1ZFFEIlsI6xBeYCvHzu4QawrW1hYpvcEJiRKec+cfovWtbzHJS7ypmaFeBMuI+j2QCEQ9zpx1c1C4qkzoFEodDLYbDyEV5Avh2sUc0DeuCZmC0lThXkmFqMcJ9TP+oOqnxzmsACs92oMvSvLtPV59/TUeffQRuos9RKw4+/BD2LrRhLFIXY+tSDFzDtmK6CR91GRCWVVMxiM6nTaLi4tUVTWXKJvK4kXMbJYjfERVOCKVMC6mdLIewocRUZkSZxybt+9QTnPiOMFUFVcvXyGLE9I0xTYNo7r5hbhvrvyRq63/8D5sjpwnHw65fvkSxWQGpWGaT+nEKc4aHr1wgZVOn5u3bqGzDC/VAybJX0E1IQ/UQEeSaDaOrqLf36KVZXWwXJgn8iIPoWdaUdiS3d1dvPcMBgO63S7D4YhufwEFDEcjoigiigNAX1UVWmu6WnNk7QhFFYL4rDEgNTIKP18rifGOdivj4Ycfpmj2AM5TloYoCmFr4TdyYSVpplVx8Oe1aze4s7nD2toJjp88y9raMX7wxtvsDAbsbm8z3d9k+cgyGydO0Ol2aLfaFIVBCE0aa4zJ8d6glKYscpIsIcvazPIZcRQYK51uh06nQxJFZO1WHTzg8FiSNOL8uZO89cp3eP/aFT77xc+ycXSB7a1dVvo9srZC/AX66z/Pvmx/2frY+yaDxM7bmCRdwbmS73/vPcYD+LWvPEQ5vs37F4esHjtGp32ESLWCZ7N3KJlgTEEkQeo2rsqRWlDJNto4KIdYoBhMGUxnbBaLtDtTvv/179DLVnji6aeDGsuXICWz7SuQxLz4ua/x8je+yfnHn2LjzNNECxO2diZkHce1S5f5+it3efjsGt4KFldW2N2+zBPPPENnYR0hS3ASYTXFZMDVqx9w+/qH/NKv/wbOm1oV46mqgtlswN5gn8GwoDKCnb0hW7t7dBfWuHbtDkv9NlQ5rVaElx5kjBICY6HfS4lUygcfvMPe7h69/gZWlMS6j1Y1e1XCXI9kK0ytOgsKrgf106rDZ4RwNlIcWV7jq90OcTfm9z54iVk54xcuPEeeW7Zv7fPwiQu0iJAE+awjNLkm4zGlt0zLgtlsRqfdZiAl/X4/BLUWJQJopSnepxRFwd5sRlGVgflsLVKCj6B/pE9vuY0yhrgSzIShdCB9xKljJ3n44Ud479Xr2NzgK4EvPJGPOLl0nL/91d/kZH+dVMcY74OAxThELOd7jLIqQ9hVGXz+p7MZZWXxPlyPtw5hDE7GzKqKylqkrcjLnK2tIbMqJ9HRv/f+Pqj/sApmYmFcvvbSS3z42husnDnBuXPnGA9HjAZDvPMILQ6BoALn/PzfDSO0Ies0hCUpJUkS1YGUBz8xjhPW19cZDHYZjUY1ABtIWFKqoKpptciyjDRtAqei2hZA1V6qCg9zK7OG0NPI/rXWLC0tzZmrD4oaAK3/Lj1WBNBTKsV7b7xBsblDqzDEG2t86rd+neVHzjFJY0oFLa2gdGggFw5hava5BOsDWauSIVhaCIH2CuEalapCuIY92ZDXavWlczgF1lh0FkLFJAKTxnTOnOfRv/mbvP31P6S6cZ2VvGL6wW1uTwbEaUJfpojCEXtBhSN2DkmwINHCYsYj7rz5FqdPHafqKCQOXxniOEFLTRClOqSvsaza01dHYWwJH8hXQiu090jjQQm0E/hbu6ROcasa01pt401JuXubc6ZkRoYQikhoLPYge1UGkBRfe7J6WRPyqhpjuNf3t/FvbdaOQHboYI3h7uY23/72d3ns0Ue5cOEC1NcT7u3H+ah+3Hnskzmj/QWewPtZj8wBnbmHKtxzuPzxoNqhj3vPLJ+xuroCQKQ1M2OZGcf1a9dIpGNqBUqGrRA6QVEDO7MZ1zc3uXX7Fp1uP9C5dUSRl7RbLSbjMUsry5jKcPLECVZWV/joyk0mOztYG9eSfBnYb9Ziq4oqz6mKAuEbXlvtaVoDxa6+RiHCBB0rQUtJumnGYivj+Po6T57cYGN9mf5CFyEcUiWMKkV/c8COg6Qbs9JLDt2Cpit+H9BcM12bsXIPi6Fhqs7/78fXnKV6CDw7YKweVNN5cwJiFWNrEEt6kKruPCDmoUM6Ovh9q6qav6aUspaFf7IHLSFEuEbhsYd8YUPnowZY3UFHbx7QdB8QfY9cqeatHh738/IHZuQOR9LqcvTkGTZfCT4lzfPhAYlCBMMnTFnhHKRJSlHm6EghCCCkELUJixCgZFAfq+Cp51yJkh6lgw+gVPqAHSoEa2trbG9vU1YF3h4Aqo33aehyijmQHDDXsMh4CEBCDaYKaTBW1TL/wOZtQl1UlCB1NO/eWeuJkpgoSkO3yAkKa9je2sNVGcIZ7u5uMs5nnDy6SrfX4dlPPcN4VvDGa29gS3sIQAlq68bEvbGbKMsqhHKF1mB9XQ5XM0ybko1hPASwDw4A80PyG1kz7DnEzv65r0NzHt6TD0d0hGZnMObU0ePYomTrxm28ABFFFHnBQq/PwsICIomxgEmCpMUKhTQVZVUxHA4YDvcZjYZsbGywuLjI3t4eXghUFLG/N6Esqvp5jII3kbFMpjlFUbC4sAhKsL+zj5mWJCrBWs/dzS10fQARkUZn6Zwf6JzHS+as5Xsu87AEmwZE9uzducu1ixeZ7Q+RzrO3u0d7uc9glpOlCdPxGK0U5x95hKMnT9S2EQ/w1J92OV83bH0jlZOcPXsC8advY22F1hllXmKtZTad0mm3ybIWnU6HyXRKVVV0ux3Ksgwgay3zN3V4nvONtC54OkulsMYF2Zx3WGfCAWZmUFFEmqTEdNjcustkf5/uUthTlEXjy96qiW2BqfJx3GfvIWt3aLe7GC9RUUpvcYHfPv8oKop4+40fcvmDD7h5ZxftwZYzjI6oSkNZTdjZ2SRrBf/hdrcHQrK/P0BHMePpjIV+hFIa5x1Xrl5msbdAq9MmzVKiWFNUglai+eiDdxG+4j/7n/8dljd6aKXYvzvkow+vc+GxUyRxVHfTfrK6f4/3AFz9D6+wJlmGgz02r/2A917/Fm9d/ICjJ84y3h/w3rSglIKje/uk8Uc8fO5xekvLIalcZAhhiaWgMoZ8tEvqFok7CdPhiPGOI006lNMJk6LAdFbxTtONc37vX/8+utNmfW2RTivD6QRhgOGE77z8LZ771IscOX6ayhesHVnBYUkWj6N6d1DZS/zCf/JFkn6GtMvcunaD0kt00kJi8GhcXlFNJ7zyytc5/9BR1tePMhkOCJZRlqLMmc7G3LlzhzyvmMwKJrOSN955j1/72te4/O5rrPZjcA6VxFilcU4gnUdIzdlz53nrtTe5duMuR46sEiuDVxG3b92h2+2AirDOhoaDqLDOMB4NOVgaHozZn1bZQ8GnzjliH2EltEXK548+TXt5mR+88T3++Pvf5vTKGZ4++RgtJ4mIsMLjlAgqgcmIoip468038cDjjz8e2E0yWKJIKcnSFLynKku898RRxPGNo1TCYqclAoGJHZf2b/LHF1/iES7w9LELdJyGSiHxVHZGJ8r4e1/7z4hVwr/+d/89u26M9YJj3WP83V/8Tzm7cpo0SrCY8KwhsRbwYftonGVW29Xls5zheMR0NsWasHcO4AXgwAqHIeynhXeUVclHt64wLqcs6I+3fXtQf7kS3jcqY3Z2tnn9By9T7Y+ws5zJZMznP/95vv31bzCZzXDygGnagKmNB3sURbU0v5HvB7KTEI3sHhpCj9YJx4+f5JlnnuZP//SPw+9xyD4gjmNarTatVockSWqf1BZxHKNUYLNaW+G9RSo9B3Gbn908Y3FNIHtQB2UlaBsCdBEgvadSAryhXVrSUY5VEY/95n9C59knGHmFs4I+isgQzt0eECrYKDqPt+EcWWERQlE5GxSgMsjWIxG8cp3zCKVAqRCmLAVK1AR2L5A64u5on9uXLvHiM8+xR0XVT2k9/Rgvnl3je//iXzF440Pk/oypUkQLfdzI4JXHC4eyFdoFlbDF044inIfh1evcfe89Vl98ntznYDSREYgkCQorAYgwaQlRk47qc7T2wYbESaisRxhDO4ooxiM2P7zC6sIKndU18qu3aL1wDtFfYVoNsLKD9YLYSpzQVLKqA6Edsg6fcjacva2t5szwBmNR9X0+jEk5F3BFrQLYGiUR1lhe/uGrDIYjnnv2GVqtLFgp1OQp9XGD4GdQPxGYetgnUtay8sCKC3dHins31T+6NREHHxMNe7Xxq/N1eny4eVnWInaOwc4OHsnq0gJ7m/sgPUkcUVaWG9dvkI8n7I9mzPICtGK57uyMRmOOHz9Ot99ntD+kqkraLcnqkVVu3b3DeLzPcO8uCxfW0EIEujPQ6/Z45Pw5dra2KHG1j2KAwaQIDEYhJF5IJKAEtGLFSqfF8dUVjq+ucvLoOkeWllhfWaSzuITMYnJTsjWDi5s5F/OKxYef5pmnztJp6cAq9R/P2BWESXre4GjurT+gTjc3+35QtPm6MGjrZOSaTXNYHgD3hjEdbIIO2IyuBhHnTE5n514gh79fCHHw/f7AZ/UTLRG6fkIEQBXCvTPW1h62hqq2MDh83b7+X0NSgwY4vbeBcPieiRpg9wR5kZeCVqtDq7tA3O7gZyG5XAsPUuOtCdJ5Ecb28QsJOooopzmRFjgf4Z2orRMcxrrg9SdipPQ1DV4Fb+A62TR0SF0NYIZ7nyQx4/EI6lAU74JE9YBFXHfKnasdLDxa12Cqc/VrN3YIDQAfgHShFLpmdUEIL3Q1m1RHCXGqQWscniIvMPmMVIXJeTycsLW3hfIVD585w+NPPoUBfvd3fo//4d/8DrNRYGoLW2PJBK8XWS9W988qLjTP5iCscx5XS/lF2OXgna0lgfXTUj83zgbmdRRHOBvYuH/da06UO9R9CfPvIeCn7t0oD8JaXGVwXnDk5HFEpECBQlDmBaPhkERqqvGEnemMnZ0ddu9s0ml3yBZ6LB5bZ3d/j52dbdrtFv3FHq1OCyFgc2ubyWRKHEVMxjOM9YxGQ+IopizyYJAuPKPxkEcffQxTVQynE8b7Q1Ido7RiMh6Fj+/tkbZaEGuSVkYWxzjhcc3uuemWNg/3oaZVcyuEg927m1x69z12bt9CGEMxmzGdFtzd38GZijOnTtJut+n1e4zynJvXr7G6sYGKoyCTaRowzZxyuKl4P0vxQf25JTi0TSD4Lp9/+AzeF3jfpqoMQqi5BE5rzWg0IklTvId2p4OQklmeU1lHO4sYj8fhtaUkLwrSWhanoyikQQuBt4601SKKNLO8RGvJeDpGEUIypQhsk6rMyWdTMh1RlK5mybr6PW865c2cE8ahc5AkKU6Ehs8knyEigZtM6S90eOHF5/ncp1/EuZj3P7rC91/5Iflklzwv8QjGg31uXx/T7S9gyzywJLxCKM3K6ip7u7uUZcXy0hICQSvLiEqFKXOMbTwAZ7RT+OWv/SrHjiwhI8tgOGBpZYVXX3kDoR1PP3kq+FvW1cz1P++ywb9UHZpovG+YkC747VK7urmS3Z1ruOIOu3c/ZKGzx2//9m/w63/r72CFIBMlBRVadHBmwmy6yZ3Nywwmd1k7dgal2xivwFdI5bBmSjVR+HxMVTpmRmCiIaPREKEk7cUVPv3sl/gw/l1+8P7byO4S169f56FjbWSyCvEyk/EHfP6rnyfqP4anQogxng6/+OVf4fsv/3c8/6UXOXd+BVLJ5s4ux48c5creHk9+9ms4ErSTODdjd3CZl77xRzz7zKfY27yGMxZvg2qksgV5nrO/O2I8nLK3t8toNMEBo6Hl3/7hn/BLX32Wu9eucOr4CsPxkJX1hxgXY5yZ8fhjT3Ppw8vs7t5hqb9Cp9VFRSWTUUWr1aWyJVSSJGm8UUPj+u7WTR6d+x0z3/89qJ+87m+cVFXF/v4+cRzPz3NCKLwUVIlGG8HTySnOPb/Cdj5hIe7Ss0ElhJdY6RmNxxSzKUJJiqJgd3OLdrtNu9NFRQrpQyBVWAMESRTRbmVIAqkgL0sqX5B6RSEs2/kIOorkSIvvffgqN7bu8oULL7Cq+9iiwGPwxtPKY/7eV/42q2mH//e/+W8RnYTf/uXf4rmHnqEdtwMwgqvTpUMTwFiHrSyzcsZoOmY8njCeTJjMphRl9SMKQucdWF/nXNTKMDxbe5tMixyXhoYennlGRkPquYd/xAOPXzi0pz38sXl3OxycDwKEHWVe8Pbrb9JSMbujGbPtAdFij1Grxee+8Hm+993vMZlO53J/rTRS+hqMlXMVZwOqOmdr3ovEuQMfUyEkp0+f5tlnnyWOo2ClVpn5HjH4oYbzn440cRKHhm2coJQiinR9UWG/Gn6WnwO5cODZ2u/375H4P1CL3FsegXJgdLAvYzSjGOec+8KnWXnuSaZKEqOIlCZFUuKZRJ7YCXRhyEWQ9TsRzqHCS4TzxMgwLqylsiWllGgZQG/jw/maWkEV1+HT2kmsFPzhH/8J3/rd3+PM//7/wPLZk0yEpeqliNYKj//23+LK0ktUb1+iv3MDCQjlEXgMJVaWQf0kErxwwb6usrSTFtfeeJ3OqaOk/QWESkBrrA5WOEopNB5HHZQloKrP2QIBTgTbSgdKRmipmY5G3NrfYeXYWW5d/4j4+x/w0ZtvUrW6nP2S48RvHCPXoApLIhW5DOrEBgsK99/XJLygDJ53NQ7hXg0LO+BHTcZLIH5ZHCJSxGRcv36Dcjbl2U89T6vXJYJgS9Cc5w6dbQ/W9YP6q54z/wIy/7AwCClrk2WPVDXghw/gDYcmftEAqAdAoSd0CeZXWCd4B4aSJlIxzgq0DIskOmV5YQl7Zx+Pg9mUQRRzc3ObZ596gRube9za2mJnOGQ0vcjK0gpxHLO1u8PNO7eZjaYsLS4iI8mfff3fce3WLbqdNs8//jDnTq0HQEqEwRRFmhdeeJYin/DS976HtZ6g/hU1QzRchkSQKsVSlnBsscOZ9VVOnTjKxsYGSysrocPU7aLa3eABO8355kebvLFVUDno6gShk3qSbo5dB74oHLpbkhoUOCwtFyIgSYRJvFk8pBD3AJsHA1ogRO172cCGh8DBechS3VG+B2S8D0i01uJsAwBD479x/3+uBlU/6QOYA7zwGGuQOnj6BnKiqBmZYg4w+gZQrb/X+xBFcLABEnPmwv3Asz0MpgqJEw4pEtppC6FT4m6P0g6QOiGSIIiAGT5LsCVMB0O6vT6zPEe5Aik1iAirwFtHZQ3FdMZEgXRtWq0UvMQYsHi0rsEGGzrblTE0RuYLC4tsbt7F2iJsUF14zxqbhvBc1un2tmYSC+qJC5xwYeR58C7cP2sKpFLB41QHLxjpHY2bgPMeLTSRTMK4M4ZiOsaanF4nJUladHpLLK0tceLkMTqtVr0xgb//W7/FdH/I//i7vx82DEqB88F7KD7wYQ6qbVf/KQCFMSGpWiqFcUGiGwmBqse/VjrIXH0IuPL1pggX4IKqMhjj5kmBf/0rrGL3CiNkfZAM87iTAmUd1WjCeGeHrcmEpdUjlFND1sqCV2MrJmqHxcyYgihWnHDnGG/vsr+1S7a0SLa2iqsqMgnGW9ZXV2h3u0xmOSrSpGmbrc0dnPVEsZ4DkcLBcDjkxvWbLC0vc+3G9eBpOR7Ta3VAQu5L7ty5hdsfMxsNSXpdRBLR63UZ7e+zMFvBd4JVRuyaq27WMI+VKhxIvEdZz2xnj0tvvsV4dxdXVWzubXFn8y7SR/Wm1XHx0kc8cv4CQinyImdw/QYvT2Z87pd/CSPFPNxMeuZrnxM1y5q/ENnvr3193Gb/Hr/wumEY3jiJspKT6yucO3OE7XHM/mAG1hAnMXEUsbO7S5ZmCCHI85y03WY8nVB5T9puMZlO6PV7dLod7ty5g5CSwXDA2toa3V6XyWRSW4lY2u0Og9Ew+O9ZH+RGNgRM3rqzw/DWTU4eO0IcafIKyurAcmDelBOHvRhrtr2EXr9DmiUYVxHpCKQk1gJbGgovEUlEpOHJx87w4nOPMJtNcEIwGAyZTGaUheH119/ijTffwTlQcUpvYRFd7rHSaVNWhtiP8N5z7f3L4C2YioceOcevfu2XSSLN8nJG1pJYZ7AmrHG7w20Gkwmj4WTehPzLvJcP6sdUs/cV9UgREvyMcrTN1vWLDLc/QLgJKslpJQpERiYc1oAXGbHSeKeQSUwSLbGmYva3brFz7SJrxyRSLSJ1hCMnwmPLISrS2NEAa0pM0iJVmrMbKWfXTnJ0fQX99JN0TzzGo8/9Lf7oX/3X7H77FZ7/9NfQfoSTjp07Q071cySBDeqoyHpdHrnwPDc+egNXtClGBU99+lF2rl7k1NnjRMkCkUrAT5mOd7i7fZlxVXHq7DPcvXkDaTX4irLKmeQlg8GYO7fusLc5ZDyoKKoS70DpjGmV8/6VSzx08gy3Nm9xamONweYmrYU2q2fPMhgMGA/2mA5nHDtxBIul01mgmO6gVUTpQWsJwmKcRPsIiWNn53a9iQ2f8w/MKn9sfVxD5X6VXOMZKqUkjmMgWH+pRCOsY1YUjKdT0iylpfqcaS8GgwUVQKLRZMxwOKayhiiOSeIY5QS/8NVfmINFWZzUGQ1BveU8OGuwXuCQqCgh1RE91YKy4vr+Td7buYLLDDYfcn5lEWcqvv7mS3zxoedYlBneB6UWlSFG82sv/hrr6Qbbg10unL9AK0qRqDpoMOyLhRQoLSnK0AiY5DOms5zBeMxwNKKoynBEa85uzkMd1CI9JOFQQOUUlbXMyhFvvPUaZ79wEukI/u0/cvR/0JH9uPqROyUCWaQBUXEeJwIg+vZLr3Dz4iVWVo/gdidMr2/R7y8ya7dJl1JefOEFXnn5ZcaTCTLWSKUQ9Z5OKDV/Pw/OzvIeubIQwWP1zKnTPPHk46RpeA6stUgVwq+11nQ6HaI4IkkzWu0MpTQgcDbsT70OGAs1Kcg6V7Og3Tx0CgLI+kDi/6OlXFDXai+oRGCIxM5RReBKi+/1OfELn0fEEdoLpA4EuVKCVQJVGaQPFn1CCZSQ4ATKe5SQRFKDEpia0CSiAKDig5cyyBp/COPGCUJuixI44djoL7DUW8CqCKUSYmHRKqLyguzYCR7+9SU2T7zOh//8n9GtSlCSxEpiH/J5ICEpE8oOFG5Cu7JoZ1C7e7z6r/41stNlceMUT33mc3SiGJWG5pZqtqS4YEWgZLAsIZzRhFfhzK9jYi1ZXF2CL71AenyN0+dP0Dn2MKvXb7E1HBNFGaLyaKlAWRIlSKwiRKscts1rwNRwbvcu4F5NEHgY03I+tkHiHFhj6nObwwuJRZAIyfbWFj945fs88sxzbCwvE9cYzOE1qsErP+m58i/+FAbUDzi0uPqmW3bA5BOiwYsPHZIQSN90IgnAnAQtY8rSYnVJq9Wm3WpRlTnjPMeUFcpKhNfoRDOajNkvRuwMbnP99jZFWbHSX+CRJx+nt9Bn89Ydrnz0EdPRhK2dbV7e22U6m6KcpbLw9C//Ei+++GmcdnilsFYgXeg6dTodHjp7ljffeIPRaDJfPJtrFQIiCUudlDNHVjm7vsaJtVWOHjvK0pFV0k4r0LoThfQzrt2d8G/f2+HN7X2sK1naWOehkw8jzATlwApZn7cOMSRrMK9JXYd6s+LmyNU9b8fh9Mx5cv0hQPSgHCGh6V6rBmttzV48kL7Pv6P+mYcB1Sa06PDH7wddP/7n/9VXY8xtTDW/h1KGcdlclfe1z+d9fqHeHfJGrZ9DIZkzbaEGRRqGqg8QrawXPB0rkjgm0ookSfBRQhwnRAKMA6I2kzLGxC0eefJToCSuqoi1nlsFYGuGGwEMHI/GZEmMMRrvg7w/TUL3SxAW8sQklFXF9s4W4/GYoigwxlCUxfxA77zFmGouXUEIXBMU1qTqQgCEACFqdicC5wkhRCpG1r+rEMH42RsfPNvqcWydBXR9fxxVWc4ntziOOHlijayV1syzAJakacYjj1zgf/zdfwtS45ycs94PWjAHYL73tR9m7X80//1rkNtXFalSSKUoyzJ0EhHEURTeYwcOS1l7/P4op/uvex1ccZB7NE2PujlQleTTCZHwwWNaS7b3d5iUJRvJBp2khfeBuW+1REQBQPfWEK8usLSygDOW4eYmIpKsr54IMikBo/GEyXiKNZ4rl6+x1F9GKsXe/n6Q6mlPWRRsb29z9OhRVldX2dvfJ4ljkl6fmbfECibbA2Re8f477+NjjdWKNM2IWy0uX/qI0fYuj3/h07hWQq2vweHAhXAXbRsWumc6HHHp/Xexkyk+LynHM65fvoGTAiFCIFsYR57hcIhAUJUlWavFrZs3uHv7NkvHjgaadN2AcfNu08/sTf6PvBoPdofHIDzEkeL4sXVuvXEjMDSQmKqi1cro9XoYY4MfaqTZ398jSdIwfvb2McYwm80OAhatZXl5mcFgQLvdpixL0jSmlWVBploZOp0O4/EYTYSUClNZ1lbXOLPaJ1KW23sznK34uDf5Xt7QQSO6026zurrK5tY4oOvWoaWglSbBx9mHYERXQe4N7SxFJYpuNzBWpFA899xjB4cvQWDKGkuVl/XXhKG30OshRFAdVKasm6UaIwzOhMA97wTOGASeSCuu37iGEI+GZsADRupPtcLdVPVez+NR4CzDrde59M4fEZcT+lkKUUrlM4SyKDlFu5Q4yRiXMzQpUs7wlcbTw2SS/ppgcOc2t6+/z/qpcyTiKD7v4nxOmQbAJ+u2MYOScjomThJaosJRcPfSdUbDgpNnvsLqkYeoXML+zphLV97h7IVHSUWCHW5iyiFR0kE7hfUGojaLJ54hbS8jx1PSziLXPniN6aTikcceRcQRQhRYUzLYvsF7r73CCy9+lSRJg/+0yQEf5P3TMVvbu+zuTRiOJuTlLARXWIFWlixL2b07YmNxwvGHznLjxjWOLLZop4KW7vP6Oy+zvz9g9egGzlfEOmJnaw+kwlAipSLwgg7WfiQMhvvc8+x+8mev/yjr/n39YSBV1GcoCHvG/f19VlaOBOmyVIwGQ65fvcbiwgJLS0u0Wy1kbUsRGHkRxocQnziKSaKYVqsVgIAoInZQaU0u6vOFC+nZIk6J0jSwA314vY92rjEQE1Y6bW6NNrmxs0m6uspyusjudMS1G9foHztfH4tkrcQyCAkvPv0pSlsxmU6QTgc1phBIpWhnGXGSYI1hPBpRFAWzomAymzCdzciLHO8PA2wHp2CJCinZWUqiEkrtGOUzKlnx+9/8Az77+Gc4trIRVov5fb639f2gfrQ+7tEV9RlHCokSMBoMefOll8mnM1aeXGX18Ye5+8N3mbx/jRNJixmKzuoSn3rxRV763vcorcMLgjqOA9gjjoPkPxDKDsDUKApejkePHuWZ554lSdM5Qbaqqrm0v5H0Z2mLOEmCR6SWgcHaeK7WwJyUorarI1xHLe9v/FIPP28HStQHY0XMmd0HhDUPCBPu0xPPP0e6uswUReIV0keUCKxzRAgSF964qTdUO2NmowmT8YQyL1BeBG/jWNPqd1ldXqHb7+MijXeO0oSoaduoga3DOUGkJUYDWvPcC5/i8XPnWTt+FJREC4FSAu2hchaxIDny+DmG/8yQlJZYSqzyKEAagUIjrMUWBhdLfKRgXLBqJR/dvca+99y5eJ1up8up55+iY9voVhsbaazSaOtRCCIbrP1wDu8sTku8ViSRRgtHtLjA8ReeBSXoxYrykdNsDGYslZZpq4NXkm7u8Vowi0BZiSur+QnTORPej0PKvSaADQ7WkkBqE3UzIqifpVJ469DOI7wDoSgjjXCGvd093nzpB7gnnuDMsY1AlTuEgx3MCJ/sov4Tg6kHD2lAlEPaSABc5l9DAzqKmsh5mJ0aytWfa3KVGrBKxxovPEU1Izag04Qcg4wFWoFBUZQVSM/OzpDsylUmuUcgmQ0H3L56lZdfusOtGzcY7ewzmUwohQeliIREJTFJO2VrMMQpTRxBaW1Axb0BUYNxzoWBJiXGBXYR9TUoKWgnio3FHmfWVjm5usLq0iqdzgJR2kYmKSiw5YRR5fnD96/zw9slFs2Jsye48Ph5Tq50mV1+G+3BqYNF8zDAOQf1GksFUTMr3UHC8ccxPw8DmQf/uQMG6X0A4uEgqvv/O5xc7+pAqgC8Hsj5D0/gDcDaeCb9LKoxCm+kMniPqtMHhHG1F00NdtTkJ+/uvR8BZKnDxWpP0eYu2/rzYXEN4WUCT6QUUZSEIBQpg3xDJ0ip6lQ/xerJU7z49KfwskvS6jOaDsEbEBovgj16Wi+upqzQ7TamKrHWUVUGXfukxlrfEzAlpSSJY9bX1hlPRrz11lvM8oLKlrVvapCvhvfvIBzMGINUEZHWtS1DfS+EQKjQYYWaICbknAXtEKg6tCE8wx4hVbjPPkzKAo/WOshta4Z0EmchTbAMtH0vPKWpmBQlQkniJCGfWbz1dbK6wxhbvz+HTeGb58GHe631Pezh5toaI2sIkiwlJVIJrHHk1syfA6WDZOznq8I9bch/EDYgAtDGMtjd5+6N6/Q6wRx/fXUFFcfhHpdl6MIqhfQSpRVWB49Ub8GMJ8wGQ/Y2t+gs9SmswQlBpDXddp/h3oTtrR1WFlcYDobM8pxWv0McxxhjGI/HSKVCSntVkSYJs+mMKIsR1jGcTNm+dZvtS9exQqDj4H2pWim3b99CGoduDxhsbHDk0YeQ1mKtoXIVwjpszVLwNmx6RtvbmHzC/mCXnbvb7O0PgQCeGTutDdRhOp0xHk9QQtHpdmm121y7cYPB/j6rx4+FYELv603BIVubB/UfVvP9UFhLpPScOH6Eb798ESk6oSHkHNaEULpuN/jNaa1x1gTmUJHjrCWNE/LpDKU13XYASb11CA9JFFPpktlkSi4k0+mMfr9f+zG70HkXiqKYIZXgrffe5szJdZSSJHHKbDYDws8+kBY1AVSh6VN33shaGYv9Be7cHeKsI1KaLEmReIRwxFGEVkE+mMZJ8OkqA3Neeo8TFmdLVJqiIx3malugvSfLWuE+1XOjKSuE8NjSYG1tRYPDSxuaqt7ivcQUYV1YO7LCnduXgl1MPZU+AFJ/uuUw9coZ1lYz3ubyy3+AmX1AHGcY12Va7uKidmBTkhOJFi4qiGOBx+KrHGUkidJY4YnSLumyZjS+y907H7G0oUjjZcKaaVE6pvKeTmc1MJ2dIxIJudOYcp+7d3d46BfOM53e4rMvfpb9x05S3P0hZT5CZR4/uYkvJKRBeqqEo0KBj2gvrEOSc+Wtb9Hq9jj5+PPIqINSAucrpqM9bl99j5XFZY5tnMTnFYPRmCvXP2BhcRFnLZPJiNF4wnCYszcY1977gBK0VEQrkSwst5lOZty9e4uz586QzwyrR9fYvfYh3pQsLC7jgMoU9DtddneGxN0EpQXKyRBWST3+8bXXYW16Cc0m52cwIv7jqB83D4TAnfo8WDf3nYOdnR3effddrLU88WTE2toaURRx7OhRFno9jDFsbW6yIyWrq6sYYxiNAys1SVNEDURqpebzuiUo3oQL4TDOe6z1xDomSzpURYF1JVZ6Jr4gaUUMtseUeUnbp3TbXb5x5118YXhu4RFOn1wLQan14d57wIK3nspXJHFEki2SVxVKKpIsJa73QJPplFleUJZlDabOGE/G5GWBr9UIjbXUYVBVeCASJElMqhOGsyG70yGFL7iydYWLNy+xsbpOMN06vMdmTvC4v1X3oH60DrIRgt2XM5aL775HmRdM7+wwXt9l5dRxTj/9GFuvvc/O6xdZSZ+g6GR0+j2++Itf5Xvf/A7lrIA4hE8qrUM2g/cIVbNeawqZqtW6x48f59nnnyNOU8JnwzmxKMs6LErRbnfodLpzr3YpAxFAK03pgjVECMYMlkVKhCBMX+8vtdZEUYQQglarNX8+HtR9VZ9N8SKobXxghd++eYdzTz9FFWl07XUqnKdyBiFAI1BFxY1bN7h4/Qp2ZxKwAx3UcxES7aGQnlJ4TF7Qa3c5cuIY68ePsXr8KLqV4LUkr0LTWnhBYQukjImEYu3IEcTSMlESU3hoEF+dxpiiRE9zdt67zOrMUe0P2G5npHHEghFIK/CpxgmPlhbhFNIrnJc4U5FkmniaI4czrn/nZfRCwsqxY1StDmm3i0xiHJJSCCrjsKYik1GwT4kUIonwdWiURWLTCKSi8pYq7lIsJQjv0STYKvhBezzSOnyi8aYC67DWzJsMZVnNcSGl9BwTcM7NsSzrDpoEUSSpyiqAxCLGFyVCQqVASoUuDIPJlO++/DLCPM2pU6fCcykOEMd7hsEnNOT+AjL/5u8N8Bf+3cCmHuYy6Xu/72Ch9YCR4XAkfO134IM5bWUrpIRYalKtySvHqBhz/PgK8aVrlLlHSId1gg8u38ZOHTenQ6rKkKiE27dusbmzx3g6Q+sY1e7RXegidQylI0ljuqsrmLTD5t6I9mqPWEgqZ3HCYiUBOfWWWElK1ZjnBqpHw95oJQkrvQ7rC12Wez263R46SVBRhNQaJx156fnOpW2+c3ULK2NOHTvJ8uo61tYgaZRQ+OoAam5YTIIgQa8/1uTLh4VUzIHLw0Bmc5/vAWN9QPcbhs9hJunh9+bjmKSHAdVmUWq+zns/D586/HqHf74xAZj6WQCq4Z4cOtTWPjZSKoyw4Guv0Dkj0dfgaJhwZXPPhcBLOfc2asaxau65DyC4FCGd0UtNu9NDKo2OFVGcYuKwKTTes7x+lvPPv8CegU6nh/OeLJWUlcSJCOElaaooS8NkMqXIc3q9HmkSUxU5QgaKuzWe0pWhm19ViBrwhuBh3Gq1OHbsGKPREGODV2ywNAj+p87ZOWhqraNRE3khUFIjtJh3Whvg0vkQ6iZEQ+EPl9/4mFpXM6l1SAfEWQShoyuUquWylrIoGQ3GSCVZXOwxHg+4cvUS0zxHO8sLzzzFN7/1Evgg8xei3rhIESbZOhwNws/2IjBjGzZy8zlbB3957zHOIj04Eyb2wLYNF93IZbxzP0fnqEPoKdSHyLDhlFB72QU20M2bN3CrK7hOl+lwgilKpqMx48GgbqZANcuJex2e/PxnuL2zSTkcsX35Gr1+j4efeozO8iJFWTAejJFZxO5gAEbS7/QZjsa02i1UFMJ/pJToOGFnb5czJ08F/0khmOV56ODjqYqS/Vt3GdzdYXcwhFijoogszdi5u4lKY+JJSauXM9nZoRqvk8YRsZQoHaNrk99SWMoKhHOsHT1Cp9cmaqdcunOT7ekgNFNsCImwLvhTFWXJ1tY2ywtLZGnGYG/AiWPHeOk73+X0+fOoNMbSrIsPaE5/2QrWN47GPFV4z8MPnUQrwTQvabf6mKqi3+thatb6dDab+/WZqqLdbmOSmOFwXDNJHMaaeVpumqaMx2OSJEGJ4Gk6zXPG4zFRFNHr9uZrShxlxFrwyKOP0W1F3Nge44HZbFr/vk3JhhbRXMjB+qEEKytLmLc+JIlaRDqq2fWCNInq3yGiKA7WXKEzkjgOYRTO46oQ0lIZSxRrbP11Vk7mqahKSZSKqWxgfNmGKWBcbaEgQtOq6eV6KIqch889hFJ6vob/++oBC+YnrRp0Egq8qJuwhjsffZvRle/yyg+v0Vu7gJebXL11h7cuXscZReIkJ9ZS1o60GA1ylvopC52cWCXoTNPud+h2lmilHWQkGA93KN0rbBx7hDTZQBKDyqlKhzcpUdzFywKTl8gopRhLOkfP0V8+ye/883/MExdOkkWn2br0XZaXSybTHK0V16+/y8MLn8PFLfACxRjpEorpiA/e/nekUY+l0w9jkx6x1HhvKE3FePcGVz94n0998RcoZwWyKjh6+mEqM8OaDlVVMBjus7M7YGd3wqww5GUBQpIIRdKNOXKkQ7/TQccO7ydc+miTr3z1F3nntX9HPgq2HKUTlLZkbWWV7bvbpK0MoeqsABGUNnOpoXBz1lf4wPz/HtSPqR9lo4ZzQp4X831Xc14Yj8c45zh9+jTtdhtPYOaFRpGi3WojpCBNU+7evcvFDz6g3++zublJWVWcPfcwWjb2ZWE/Z6wFKTAqnJiUDYd3JxRJ2qIqDMV0iqVCpgo3K+hWMaePnOKVG++zWW5yeXqD/ckejyyf50uPfomu7NEkTkMDwAVQw5bBizCKNe2sRZqmwY81z5nNZhSVoawqirKkyHPyIicvK7wLxIOmDqumAiHBo+KILMqIo5hr2ze5fOcqFsvUzfjozmWeyp9gIW2ja2/lsJOoFWPeH3ip/tW+5X89qiaJfPje+7z/1jt0Vhbx2yPG719jodujfWyFNSnZff0Dbr36DsfVEyAioqUen/3cZ3n79bfY3tvFqTpATAUlnqrJHUoGX1UlJRvr6zz51JOkWTZXJjWWd0VZgBBEUUIUxcF/vVb+CREUNt4ekIGstRhjEYq59FnVgT7NvkBKSbvdJq2B2/uDk3++qyaLEJ6Zhmbji5LReIpe6FGpkIVjpAPniAT40rC/s8v1jy6zv7OLlIK2TtGRRMURQkHiJbFQFAqmwpL0JMJYxlubfLC7xXvvvE53ZYmjp07QX10hbQe/ZackSkiEDZ0RFWkq4am8Q9VEpQqPjyXxNGf61nsc9Z6RltwaTahaLTpJTFd4ymlOqYN1lK2gFIIygomqsNJBIhBOY8ZTLr/yKqPtLdZWj9A9skza7ZCqGB+pkOGiI2wCSZQQK4moyV1B/m8D1AEIK2ibiFkscc6gbGgoGOGRzqFKh8oUPlJUziBVwEYqE/AK5xqsys2Z+9bacG/q/BZnCTiMlKgoRjiw0mJkRVIJxNTjtMRR72mt5dXXXkNqzckTJ+b4+WECHHxyc+VfCExtgFQOg3gE2XTDevTzrxdztJjmkClA+QYsJPjDIMknObdu3ObIygpKGnIZs7c/IlGGsS9oR4qiCIubEYLKe27cvYuOQGnNLJ+wsz/EyZiku0S2sITu9Tjx8MO0uz2uXbrKZDgg7a8zKaZcv7tLKxa0YkUWx2hXG+cqgTeWNIoopcQpH/zQhEDIIBVuxxmL7S79Vps0SdFxTJQmc888ITS3pvBHr3/IZGpYXm3RXu3gXIkvCkzZIW0vkM82aYkDsLQxJKf++2FE/fBEeRg0bTYwzXtxmOZ/uNl+D536vtcRHHTyDgOrhwOlDn//YSDVWBMYL87XQPAhub8/8Mr4pKosy5p14OeSCFmDg0EafwAWeu/rxVDUkl7ZPImBbq5kDaKD8/bgXhE24hLAB7ariDLWjh7HuhAApeIEoVM8kLXanHzkSd69eJm9POfJJ9rsb2+zsJjR6vSxLkILSVnMcM7R6bSpqiown6IYUxZUZTXvtBrh2Ly7ye/+zu/y3KeeZ2FhAWMqer0eR9ZWWVtbCyEOg4jpbEo+mwW2B3Zu0VCZCodHHewlkdGBZKU5aFTGECmBizWmllQ1i5SnBjVrA4XKGlIR15Lv4N+q42CcXhnDzs4uu/tbdFeWqZxj8+Y1bl+/TjGb0ltc5KlzD3Hn5i3ev3QZpESplOA9GOqwv60QzfukUTIc/qWU9X1TB2PbgzHhInXteSRkmOQbsNhY83PZ6/fNJOxrH9Gq5Nb1qyQWep0WX/rSl2tPbIEnMJWdqZiMR9iqpBNn2MowzKdk7YwjYgXaHY52F7DC0V3shnmgNCwtLDGblMymRXi2dESn22E0GWOMIVUxZVlxc/MunayFjDS2TrptmgVlXnD72nWmt3a4ffM2JlZEWpNkGdu7O7TSjJ2bt1n2mtbDD3HqsfPQijBKIl3wBa5cCaXh7hvvEynN+iOnKWNJ0m9z8txpvhppXvvBq7z7zvsYKpx3wevKOayEwXDI5tYWw+GQtNWm2+9hihJFPRfLGp2q1zrJvOn8oP7CJWCe0xn8w5cX+rSSiNHEkRc5SZyEZ1+rkOqpFaU1FEWBc27O4kiSmDzPiWTEYLBPHCf1XNup14zQZCmKIjBZs6xeBx1VVZJlGe0soyxmJElKkkZoHWHKGXme179ho1i4by5pNndCYK1hcWGBdpaSRGHrFSxZQEcSlGRW5AhCIqqOotoL3aJ1YNlrHawMRNgoEOvAUnHCYp0jjqO5TUoko3uA0fA8a7x3SBXN50mlNVevX+GxJ75KZd1cnvigfopVM5Wb2+qdwRS32bqxQz6xRD7lwiNP8K0f/AsGuyXHTy5y+YMtZpMJd25eYWNRgcro9NZQC8vsjPcpZp7u0aPIhTUcMWp6l1s3Xmaw/V2efvzLiNY6TgRrKitmRFogvGVSFaQyppjscezxL3D3xnt87nNPMZwWrHSWeHVrwomjE2Z2hlKKrt/GVFNE3KFCkPgCzC7f/cN/zdLyAqee/SwuahPLFt6NcM5TTie88/r32Fg7xcraGYrZAGlLemsb/Okf/g6/8it/kzyfsb+/z3A0oTIe50WwCvKgO9DpSTrtNp00pqxGbJy+wHJnmTsfvYWxMa4doTUUkwlZFDxdc2NIU02I29AIVScX12QEao/MOdFj3vd60ABr6s9rkjSfms1yNjc3McawsrKC1pqyDFYjGxsb83PfbBZS7htWX6Q1ZVWCECwuLYGSjCcTWq0W/Zr5KaUE56nKKqi96lBTjMFbi3WWWCiyOEM6z3A6onIl3YU2DktLd1FWM9vf4QsnL/BHN7bZvrXLI4un+XuP/Q1WWUMYj6Wan1ddfS5t9vmVtbjSo2pGbFlV5FVJUVXkZVUzUnOqqsIai3MBMFXq8L1qWLuACOeILEnpZT0qV/HaldfZz/ex3jCj4N1r7/Do9Qs8e/aJoIhAEiJhG4ZVc6DjP9qh+ldvH3NwcwSBnPTGyz9kcHuL4+fPoo8VlFfvsvX2h+jFDu2HjiKkYPuND7jz2nscQxDFEXGW8pnPfoar169z6fJlpsWUCoXWitQHgo4UAq0US0vLPPPMM3S7vXqaFzRS58rasMcQh5R1UmJqmzEpNVEU33NfGs9hZ0PgWVQDpo2vZPO93W73AXj655avMZAAhhd5yeLqCjaO6t2lx2LwWFTlufbuRa5+eDnk4siIjkyRSRzwGCWCVbMPz7HWEamUJChUKmgt9WilKYP9fdyk5MNvfZ+43eL4+YdprS7RXVlGSYl3HuMN3ksK43BSzvNu0AJblEyvX2fn1Vdp2xEq9ljVZq+7wCYlCkc3F1TkxChkljCcWcpIYxX0SemvrTPuLHDh9CMcO3aEl1/5Ptff/ZCllT6LC31IWsQLPeKFHslCn2hxgSSJkLEgUhKMQHhFJSUFJVrV2I4TZOigTJTB2cyXBmc9hYKkVHSyjIl3lGXIM2ls+YQ4OHM7H1ipTbh5ULgfZPgIIVBaExeSqZLYLCarICklYwFWe7QJCoXclrz2+uskacqR1dUazzrsofrJjbafCEwNLEnJPGCBA5akIAAUiHvBv2Y6OwACw+ekFaBEHTajydIuIklJ+4voTg9fFmxt7TEZjSmkpdopiOKMTidCY9ARdNKY5VaLXmuB7cmEa4MBOo7RWY/2ygbLp85w/KHzGOPp9XpMKo/e2iRpddjc2ua7r7zGu285jq+vceHsQ5xaWw2JaT7IsBIdkUYREIBbL0KoTao1SZTQbrWJojgMKq2QWgZpshCUVvCn73zE9tY2re4ynYUj5Mah3BRrEmwlUElKMXZkyPmmIwSeBQBIHoJ3hPBzpur9naePk/kfbIRcvcDf63/WsEebTcN8M3HIL/UwI/VH2au+9hwNHePDtgBwiKXKgdfoJ1V5kRPXh1QlBaGHCL4GhYEaWJXEkSJNwmHT1oFHQshw74VEaFmPc3DOBoamFEhRA3P1RC3xJO0lWu0elbVBfqliZNLGOcuxh85zbWuT0XTIudNnGW3dxc4MA2PJWitYwNuSOJJImTCeTtnf36fVahHFIe2xyGdUVZicer0eH314ifMXLvDkk0+xu7tLp9Piz77xDdbW1zh37iE21tdJhKBodcjzKUVZMs2nTPMZeV5Q1e+/sWY+gWkdo2o2YDOhOWsx1mCdQtpaJCf9vFMeXAEUzkGeF7RbChWpuQ+nkCGsyjsYDUfk+ZCpASzMCsN4MmN3axO8Z6Xf57knH0EIy+3bd4kjSZJEaH0gmcIzB8mFEEihAsNWqnlTQGt9z3hM4wRTVQgExtqQuKnD4crOx8XPKXLgHVIohLNUVUk7Tdi5dhtfhsZVkiXB0yfWVM4Ef+tum8hnFFUFkabfW2F/Z4fJdEKn20YtdlhdWWI6GTMdDslnBVM5w3mCf5CAoqqY5hNQgiRNeOeVN0izjOFkzKNPPI4xlt3d3bmHpXeOuzdvUe1OuHL1Gl6rILtKYvb29pGRZrK9RVI6Hn7+cX7pP/1bRIttnPCUrqIlNKYoqBRoBdOyoBzvMzEFWb9D4Q0r62usrx/hF3/1FzEYPrj4IdUwHJoiramcRyXhIN7pdpFKkcQJK4tL2NKgkgjbjHsI5/RDZ58H9ZPX/P7N4ckwL7fSmE47Y3dYECcROEdRFrTjTjigS4kQkNVgaAhmOJgTyqpidXWF0WhUsz8M/X6fsiwpS0scq7mPWWORI4RgNp2ipQypuxqyTNPtdtjfy8myrOFA1L+9P6QtCs2nZkwoKVg7skySSNrtmChWQY0Ta4qiwMch5RcfDtEhZKUKlgRS14CqQEiH8w6BxJi6geQESmiogpcqMnT4pa2bSIj64ODRKgbCniVNE+7u3SWKJaurK2Gd44HE/6ddwguk8IgGHhExSfcR4oXHOfeUYeX8ad588xXu3rjMl549w9d+9TEmE8OrL99ktH+V5WXorxzn1ev7DK7fZGF1hece+xQnn/sCnd4qWmdoCm5fPM5Lv/f/5Y3BH/Pkl34NnfRQwmFSg68CaJtlERfffg9jY55YO8affP13efqRh7h2bY8v/9JXwSpuXLzDnorZHt/h7z/1RWb5gE7UR3lFWZW89e3f4/jRY5x94teQCrwwSOcRVEgv2br2EbPpgE995ZcwJrBZUDFpV/Lk409w685tJsMR0/GU6XRGXlVMZkEmrZUgakGnlRHJCis9jz32InEEe7cv8f7FS/SPnSDrpeAqWmmMNZYPP7hM0k4RqtlDR3jh8MIhRbDFaAgL+EACkQ2R8MFw//dWs51vbJTW1tbmZwZrLfv7+5w8eaJWToUmTlVVTCaTwEzVmqIsMcaEw7aAhYVFsiyjmOUIJUnTNOxJfVCEaKWobEiBzoTGSkEpBDpOSOOYyWjEaLRHd22RO8Mdrly9xNmHzrDaO8ayX0GXE7568kXMVPKFh55jwy3jfVXb/biavXhwXpk7ynuPcxIflLoIIZjMpkxnBXlRUJoqKL8gBFLVtliHxT+B0NKc38InO3GCFpKtYo8/e+PPKGWFLCWFMbzy7g+RleLR//U52kkSfqv5uHwwQH+SanJbhBB457h58ybbW1vY4QSZVyyePcHYC0a3t7j96rucfuFpeudPkqUZm2++z83X3uWkg/4jp5nokqNnT3Hk6Bp3bt/m9q1bVEWJjCW6JudkrTZnH3qIbm/hR868AphNJhRFgdIKrfUcKG18z4UQRFE0B1s9Yd8otcTZIOuvyjJY/dSTVeMvfNgv9UEdqmb/Vf/ZkHKqsuT4uYewWoS9lQ1fJA1cfe8SN967TOIlWkVkMkUbSRUZrLNEUlPmJTKKqaxFSY9xJqgehGCyXdFKUnppxuryGmu9JUbTGfn2iDy3jIYzWovLdBa6xJGmslXIGalMYIgKha9C5sfe7Vt0xhNSKirlUSrhU1/5RX7w/uvs37hGEgsEGlVavDNEQpL7AE5GlUM6hdpYQXRSHjp9GuFKvvNv/geGW7eIZRzCrhb72IU+qttHrK6QXjiDPHUCOh4lMpyU+Nri3bsKoRwlisiA8h6HxVLbr9SBvsZWtJ0gyzK893NA1VpbK0TEPXaQUEcAzclRB5aRUSRwkcYqh/GOMg2Bg4UM6uKOFyFIWsJ4MuUHL7/C8889y/rakZogdFCfFGP7JwdTBXMZ9LzzIwRSBap7HMekcUweRTQS6Dk9slmfAOkVFR6tJY898gQnT53FRDEujUljxfjObT784SvIYoLBsusqDBqpBD2tOb7Y5tyxDVYXl7gz1dy6+AFJbxniFsnCCifOP0alE7aHEzpZB5wg0QrrLF5LvBLs7e8zq6bs7wzY35uSfvHzrC6mSCEwRUGsJFkUfJZkzbbVUpJG4UAjAGsd1nqqKniSaRRGKi5e3+atjy5jsPQX+jiVUs0sLnLMrGdcTllMoTA6yAcIEkYhGmmjn+9a/KGFXnAAejZ1eBK9F2S9j316nz1A8/X3s1GbjzdslgZgbfxTD5jIwc+y6fA3AO1hWcvhn/VJVWUqpKilEd7PGTq2ljdKFTwe4ziilaWYKpjdO+vC55QKixoCWSck2pqVqpREEK4pLKQgtQgHz6jFaDqjF7WC/2erw3Q4ImlF2DhjZ+8OJ49uUE1GjEeGLFtiobfObFIiUohjiUKytz/g9u07DEcjNjY2QqK9FFRFgZCCydjgrOWzn/8co9GIPM8py5LJBNbW1/j+979PUcx44rHHSJOUJA6SEmNKiqrLZDZhNJ6gRkNms4LSho2t8KBVhIwssqbSNSCjd3Xqng8hPt6FRMLASoVGJm69CwFQXtXdU4exDi01xgYTdW+BaYUvHZWFnemMD2/eZFQVnHCWo0dWSZ54kovdLvuTEUmkieMIUTOLm82EaDYensBIkxrhD/kL1wwI5z1RFAIEjLPzpNmiqhACIiWx1c8mLO1nW+IQo91iygKsIdGS4+fO1GB48AIWQiCcgMoynUzotDMQguEsZ+fWHaq8IEpiFtZWiZPgA7W7u8vu5hbtqMP3v/09HnviCd55733OnT9Pf3ERM51gjKXVyhiN95FCcPnyZZ5+7pnAXJlOiXUE1hFLxY3LVxlu73L5/Q9wSkKkUFHEYDhEaMVkNEEbx8rCMk984UXSxR6uqJCVwU7G7OyPuHX7FkefOEeapZz4zFNIpVCFxc2KIOvWEZ2OoqgqfuM3/yb/5B/+Y4a7w7rpYNFCkZcllbV0+32klAxHAxb6C0xGI3qdLMzdUtaM/Pr+ivle4UH9xBUOkV6IAD4FZJGFXpc00Vg7wWGJo4iizGEKi8vLxGlKWZZoHZHXjCjvQ5DadDql1WoxGo1qzzFI05TpdBrmhLIMvs319ywuLjIej4njGFtVGGNot7tYU/Du++9hrKTf65PGCYcPuvMDtAgAGgT/KY9FeM+xo+uce/g0o9GMJJVkmSKOBJFuBaYKQSGjtMa5iigKh65IR9iameJ8aCB5f7AnEKH1j/Mm+PwCxoa5W8mgCVP1YU0KhZAe7w2R8rSziC9+8TOsry0B5Sf/dv88lLC1C2LtoYtia2RJj67jR0NGu2O++6ffI0LywvOnWe51WVlOePTxZ5hNDe9/8BF/8M3XuZ0relGLjf5p1k4/TtbpoqUDP6YEkv4pWksPc+3qD3Av/QnPf/ZvIGQSGqIyoXQeWo54sc/SwuNcvXaF5d4yvujw/ItP8dYbb/Hs577GN3/vX/HKpT3+7n/5n5NPR3SWa89Ub9m88w5J1ufUE5/Hp6K27HLgRjgcZZ6jzJTBaILXCd6EAJYKiRaSEydO87u/87s8dO4ceV4yHk+YFbM6bMWjFCykLdJ2DCk8/PAjtGLNnUtvsj1wmGyROCtRLuxD0qzF3t6AYCemCMwOeaiJFfxSG26fkBrvA+CXJEnzJQ/wqj+nDs81VWVotTK8D0FTw+GQsixJkoQmtNf7wPRXUtJutdna2qLb7QZv1Kqi2+thap+8NM1oZy2MdxhnEYhASvD1WUSEfV6QT0M7C56Tw3zM2AxZPtIlzjRRukQUaxbSlFblyLOYLT/k4sUbfPHkZ1l3CzgrKNQsgG6uEQ6Eve6ckS/EnDXVhPPKOk9jOpsFEgL1PlSIe4ZNs9aHs9KBmsoBQioWex2cqvjGG9/i6u41jK9IZITFMSiGXN+6yfXbN1k+1Z0TlOZ4w39k4/N+AOOTAjWa1PCtrS3eeO310KgvK8a3tkge7dA7dwKA2d0dbv/wHY59+klWHj5JJ465/dr77L3xIeUsZ/HJh5C9wEZ96MI5zj/0EOVoysQENYvUEQsLC6weORL2zIj5+9RYjEwmE0xZkcTJnGUtpaJJMgcwpiJNU6Rq/HUDgScw+kLwmaIhAgUySRzHwUbj0Pn+QQM0lOcQoYEDtrmOI7LlPiNbIEXwDaU07N24xdUPL5M6SeJlAGAjycx7YiFqawdoRzHGW6I0wpcVLS+JkohJPoPcMFY5Q7HL3Z3toFqOYuzMkSYpambZq3bY3dtlaXWJlcVFIiEwwkHlQIQw0gjJtavXEbOcNFMQaVR3kSOPXOBrT13gj/6b/ydbgxFtkZI4RzmbgdYIrahmOWncxuxPGH94Hd9aYn865viLT/BMXPHN//afUg2H5FKj8z1aWwlVqUhbbYrLV2g/9yj9p57ALC7jsw5SKBIn6txyD9IhvQre0viApzmB9LpeGwz5zNJpt0lqUkIDqEJQX1WVR9jDY5a5XaEx5h7CnxKSdGeP2eZVxNFjVK0uwkhk5TFa4Ag2itrC/mDAKz98lS9/6Uu0sjRYN0TRPLj9r54R/xOCqZGO0VqTxMHzownykTJ0aJIkptvuMOl0KYsSrfU8zIn5ob2R3mq09GS9Drktefu9t1lYP8bIFLQiQdcYji33KVWBMTB1EbmpSBScWOjz+PENTm1sYGXEt2/cYqwyrFaIrEPvyAajvEBFkjTVWFNRVgW6ZiiGxD0PtuLZJx/nrbfe5tr1a7z9/jJffPGpEDpVVSg8iVY4AbJ+ECOtQiK4sExn0xDm48GVFbbIsVnEqDD8yQ/fZndnSLq6QtTpEBiiAlcIZmXFuJyykLZxcYfSjZC1HFTVfn6HJwCo75+obQiQYTN4CFS9N8HP4/290vzDQNH9QGcDpDavAyHd7f4Qq8OeqwdSfjcHXT/udX8Wk/uc/VzL1cOE2nhyBEm+Voo40mRpHGSSUYTzdk5DlzXTOopipApyDFeDqeH+OpI4CQseHi0EVnS4uTUh7SziEXQXl9m6vcmxjRNcvbPN6voxsjhllu8jY4lqp8yMIVKQKY0zns3tAVdvXGeWF/WmtE60FzAZj0I3Mo4ZWUdRVSwsLMx9P2/fucPKygpnzpxBKcXO7i4KKMqKKJLzBMk4isnijFacMhyNGY7//+z9WZBk2Xnnif3OchffY19zXyoza19RQAFoEiAaTRLcehlTz7RG0oxm9CDTvMhMD9KbnudFMslMI43U0vTM9HQbx5rsbja7SYIbAAKFpapQa+57ZmTsEb7f5Sx6ONcjohJFojANgCCAY5bpER4eHu5+z73nO//vv/QYDIcURY6QCq8itFd4FN4HQ24nqcDU4B982MV3KBms8pXSaC1D2JZ3KKGQSmFsSWFKJMFztV6rE0uNKUrqjRbHT51jt9tjb38PJbaZnZqiMzXNZ0+ssrW3zdUrV5CV95kxBlMFRwkRPE+zLA+AaVV8W2fAh24u3gemrQ9WILjgM0v1uxPpVFTJZH/ah/iIr/Ae7yyjQZ9mokkjjY9jQj3hKIsSfPC/jdt1VCPG+BIdadq1mE67hc1y+vmImeVFyqLk9tUbzE/N0tvu0lme4vO/+Ets7W7z6ic+wb/7oz9keWWFlZUVtJD093vgBfvdLsdOHKfZbjMajxkOh8FHiMBI3dve4dqVq8g0QWmNl5Kt7j71KGa/30VIwUJnls//xq9y+rknMeOc/sY292/dwheGKC/xseZbf/JVXnzpBfp7XeaXF6nPzhDPTdGaaiKMQ+R54O9JzQvPPcftGw+qEDgL2iNlxHAwREnFg7UHnDx9CoRm49E67eWFUBT4EGD4c/z0f/qYuFwFjmTYnSghSFPNiRMrXL/bxbqcEk+91iCtpRRlwebOTgh2kCqESyU1smxUXf80ZVkSRRFpmh5YAEyKPa1DQThp2MRxfOBHVnp/YJjfabdZWXiR92/coxj16Xa7wNLBa6+qHQ7Ps4kPd2iUJrHis5/+JDdv3WF7Z5s4VmGJ95Z6vYH3oWufaIXzEKkIayyKwMYOjY5w9YpUCPJz3iG0rnzSHVLqg7UrUGIna1qQjsZRgjEZzlkW5jqcO32CepqCN4iQj/0jPsI/W+OApIwECnAJ1oxY33qDjd11VmfmuP7+HXbuDfnSrzzL6dOLREkL4pTMGHSU8sJzT/HkU0/x5a++zeWbtzh5apmVpUW0jBAOSlvyYP0Rb77xXa7eWGOlOcuty+/QqC/wxLOvkqQxpSvw0uE8nLpwlsIJhqZgttlid5BxenGV5u46TpT8+n/yv+YftGK8rLG+3mf1/BTWl9hij6tvv84n/tZvIOsdIjUOZ6szwS9dSMa9fdY31njulVcQ2uKzAi9DKKh2Cq0btJotRsMRaVJnPB5jfYnHoWREmkYkOkJ5y4WLr1DTEcOdu1y+s4aL2qysTiNFCWaIkBFxvc1gvM7qyiJZUSKFqiyJqvNGSFyAs5BKI4SiKA29Xp/5ucYhFf7n4yNH2Nwe5ihorauGNgdM0izLqNfrVSMxKO7SNMVUNYS1lp29Pd5+9x0erj3kt37zt0Kdi0CIcH01Lii7dBRhSoOsiCW22ldY5UhUQiIjiqJkPx9wa/ce83KKeTVN1i1I4jqJauIK6OVdPrj5AWdWj9GO65gSCukRXh2EyE4s6Q4ankHiFDIBRJUhUZYo7wPjaq+LkCGBOyizqv1XBbxaW/lde3+gaAvsVUmz1mZmus3Dvfv83l/8W4Z2FN6XAK89hbd0hwPuPbzHcycvHPCQjvKsfpIw1b+KhPDXWVMLwnF75+23efDwIdPtFsVMh52NTRpTbebOHKf+1Hk2EfS3dln71ntMf+JF2ieWaTab3H/rCoPba2SbeyyePsHM2WPkFvrCkTQS5mQDFcVIHdHutAMIKo6s+Afbcc+wPyDLsg+x7oyxB5/P5P6JN7CqCAy6IvMcZaPCYdBus9k8qE9+Pj48nAA1aWoIDkDu3JaMhcV6B9ZCpBnt7nHn7fdIhhl1A5EViDjGJwpZi1FVAyf2gDHISAAOVZTUvKY/3KPZalIOh8hGytCV4CX9QQYohIooyoJaXlBvTSFkzM7GJi7Lme5MoZMYUdV6xhnqOiFttLkhPMZpjE9pnT3LeK7F9Oocr/3H/5B//U/+KTOZAekx1pEJw3g8JPEKYy0tNPu31tia6rD71BkMHZ74xMuMH27zxu/8Ho6S1BnGKqMW11mWdQYfXGbY28Xu7zD14nPUVk/ia21yBEImOFtiFXgj8EpXsTE2kJgqNXXswHpPNi5I0wD2AwfzP7CwFUZ4TJXb4j1492GP/gnRL3JjBm+9y6Ovf42Vz79G54VPhPAsKRkJC1KDCdZAUgh2dnd5++23eemll6hVgcZa6mrd4kd+4fxYYGqtViOKIhqNBo1mg6TajEw6JUmS0Gg0aLfblGVJVnUkjwJqHrDeoZzESEfaSLl36xq2gNFwRNxIGQ66lGXJK09fwHUUvijoZSGgqplq5lpt5qenUe0W79zb4HZekCc1CuuZm15gOMxInKLdSom8w/qcsdXUO9OoR2skSBIPY2eQ2vH000/wjde/zY33v8uLl05Rb7fwzqO1wjuNV6IK6RHEShJpBcIxGg0p8xxqHm8MZZZhx5oPNrd4++ZdciLa6RTWWzQZ3jfILdSyApNZNuMS6yxJnpFW4ReTFHQBRzqSFSPV+QMJAOJQbgOHC9rRwKnHmaFHAc6jt0e/NiawHp11HwJTw3O7A0m0c/6x2+9l9v11MVOTJKng5nBCTjxKwnu1lUF88KRpNkKTIE2CHymikvpXrL04Dg0EY+1Bl9FWX6dpSqR16KJ7yExMlo+wxjLKck6ePsPW2jqZcTTaU0xNLWD6PYTUlCrnzsZ9Ti6fpREnmHzE1vo+D+9uMTJ95ucXmJqdIU3Tgw7/9vYWSimOrR7DCUmel2xt7dBqNZibn2NpZRnwnDhxgqtXLyOkpD8YUBQ5canQSoQOuvekUYxudkh1QhRHKK3o94ZkWYbxkrgmSZyoWFASJTyliRDWVgFOExcneeCLq5QKHrM+zFUngheQUp7cWmId/FWFUthIUXpHI6px9vgZEid4cPcOhS1wOka0W5w4e4LF8Tw721uMRsMAjFcAOUyM/d0BwCGlREsZivzq2IeQLUOaJCAEtjiUo0kR7AeklKQVcP2zMQL14qg37rA/Jok1RTYmwuP7w8oc3IAtUVJjXUZuLQ6HiCRGCDQS6Sy6FlNPJXmRUYtStBeMByOSpMYf/sEf8anXPsX16zdYPXGcz372s2xtbxFFEXhoN1o8uHeP1lSHC5cu8fDRGh6IlKLMctY3Nlm/94CHd++j6knwvxOC8XCMVpr13h6//Etf4Fd/6QsMszGr50+hajWGDze4f/ceK+dOc3xxmW/+4R/Trtc42axx//3rZCZjf3ebRqvN9NwCrdlZ0nYLIxWynlL2epw7/wRLi9/h/oMHeCrmvZYMxyP2uvuMx2OkCl7eaw8fcv6FZ/D+kBn9k7Pl+eGPv+y6/kM7j4QneDFX3mNC4CuQ8PTpk8ivX6EwebDXEaCUQEhotZokScpwOAyhg8ZQGkOWZXQ6HTY2t6ilKVEUMR6PmeYQRPVeUBQFzWaT9fV1olhz79495uZm2d3dJY0Sms06UtW4fuMaguTguZkc72ojfuSNVLcTDyeHw7GwMMvMzHTYbDvP7dv3KHJLaUK4o7MmXLNlTKQjSgRKCaIoBE5FUVTZvhiUFpWPeZCeeeeIlEBJ0JGqruNhLQzruMF4qNcjlBZMd9rgPBIL3oB3eDQ/M5fEH9PwXgXJebhyUoxGrN+5x2yngzeaP/yDP+Mznz/D3/n1l0hrHSwai0ZpgbcG5zVJ5PnS5z/JU+ePMbu0QNMPMcOMAsugN+TelSvsPrxJmSmuPMh4fjnlT778L+mszLK4cCZICVEIo8iLHKX2aCVTPNiP+dQv/gJeClYvvsTWnbfZ3rmL9kNcMeT8k1/AUkP5gitXv8nxJy7RbMzjpKHAIp1AOYETMdgxkXBs7u4yc3wR54Z4wtwTgCslvvScOXWKqzeuM92Zpl5vsN0d4L1DRzFSQnuxxbMXXmL44BFjOeLqrW1cWmf1eA1hDZQCoS229Dza2KY9PYOSEEsZ5NkKhHRoIVFCgbABJPMSRJByDvpD5mZDi/in+HL97z2yrGB7extjDM1mk1qtRr1eO7ARS9M0AKfGYIwlig6DdCbMzkajwWA8QkWaJ596KrCctUY4j6tyGJIkrPFlWQYQ1XlwHi8FSqugNHPQ29lHRQqpYSBy3nrv65xbPsWl9mlW6x1y5+iZMW/dvsyx+RWWdAdRenJySgSxk9V+NBBTIBx+6av8Dw73WNZbcFVztNoX5GURGlr+UNnnrMUZe2Sf5PCCA8srHcccXz2FcQX/9iu/z/WN25TeINEhxEV6ZKSpN5vBPs27AAj9DZ6Xfx2Kr8nfNMZQlmXY9ynJzOlj7I9vMbh2n0Zao3PhJMc/9Txbb14mv7PB3a+8wdIrl2ieWmbpl19lcP0BvSt3ePjWB2xfv8P0iRXEXIudWOCUpNZo8sSTl0KDs2qYeqqgKn/AjWQ0CoD5JPQyBJGF/ZXWGq1UtX8KNgCmakhNrInKiq038UuFsP9pNpsfstb72dnDfP/hBYhKwOsP/vd0B33GtkBLSc1A2e+z/f5V6rt99NYeLTRzM9MYa+hnQ1zk8DoJyjTpyfoDZCPhwcN1nj/xBHanx3S9Bkqw9ugus0uLrD96wLnz5ykLg9YJTnqscwzHI1wvR3TqxDMN8t6Qzf6IdKpFa2aGJImx1lMYy8Vf+ixXb17m5vuXacqIp599Cr88TZ4qzrz0Ar+iYv78X/8bbt+6zZS3OAHaeBo6om9y4sywJCI+uPwBnaV5Tuz22M0sYs+QqDZxljNlPbqjsYlib7hLMi5Qd3PG4z66P6LxkkWeO8+41QKhiK3E+hIzsaxyVT3JpPz1FMKiVIS1lqIwRJGm0WhgraWs1F2TkLVglSjwPiiojl4pJudwKQp2Nu9SrN3n7tdf5xOnLyJa05RKIL1Al4DTOHkY8nfn7l0a9TrPPv1MUM96KvKBP5gNP6rxscDUTqtFFMd02m1ajQa1WoqMFLLyAGk1WxRFWUnCNVk2Rkp10NmTImx6nPchJdnDoBhTFhafW/rbGzRsm/5oTOwNJ1otanNtvMkxNhj/KhsORK2Z0M9L3t7MGJVhwzU/M08+GuHjmFJlFDalEbdBKqyTWK3IC8f9tR2EjJBRk1HuefnZJ/jWd7/Lvc1dtnd26HSmsdWkLIUkKXPGvqAUEbGISXQAjPpjy3Z3TFIf4K1HjVqMcs0ff+stuoN9kkag/Rtn0N4gvcF6S45kkJd896tvEgtHLYpo1VM6zTrHOjHTiaVQHuEiGkZRCEIIlqgASy8ABXICdgaAKPzsENT8EIjtD/uavuosT9inR5mpk26qfSw06hA4FRW499Gs1Q//TiDZ/ri7/amOEFJQGoOdMBOrzpS1gZsglawSnyVxEpFneQAaZZBT+KqjnKY1JiFc1oXiZvJe62loLlhnKUuHtxpfjMnHhu3dfS6sLnLqqWe5evMDTh27iC1cAMslSKt5//0POHnsHMVwyN72Gg7FsdNLrG9pFpdXieM0bHqEoixKpjstkkgQy5yoFlE6QV5a+oM+9UadRr2GlFCrpTz/8iv0+wOuX36P0hqsMxwGOTkkwThdRYp2o40UGomi1+9T5COcMziTYZFhYyMgTyOUTREVQ9f7APRbY3BWIGONkh7lJNbrSsQIUmpiJM5adBQjiJBRhEPjcTSaDRaXTxAldUqTUW81OXX2GNOdJtaUNNtTvPnmG+zubaMTjY4UUaTQUmBNALLxQR6htUTKmDzPEZ7g12o9WVGiohCgVZogzyrLUPTHSYyQgfH1UzH8498euUMIHCGISckgW1HOIUuDdw6pwzHRAkoRNgLSC6x0WOGJIx0eI0GL0LIw1mKNJZU1+sMxcUdz7Ow58uGYxrzkk60OSZxw8amLfOfb32Z1eZUP3r/KExcvsbbxkHOXzlGfbXBm5hybO5ukWiOUZmwK1na3eXD9OqOHW8hahIs1SihGeYlB0N/Z5+LLz/D3/1f/iP2bd5mbnUY0UoSQtJcWeWl1GYVn7fYdphcXGA36eARnn7xAo1HnzpVrqGaNm3duMHPtNvPnT9E5sYxC4Bp1BsMBZy+e486DhwgfmNoWzzDLcKbk2SefZFgYnFCkaVSpAwOL+ig7UVTp3T/eKL6/4cNXzZMqh96hmAQ4Lc3PkEhNUUiclGR5iRdjkiSm291ndnaOVqNBaTy97gjvJl7vDq0ESRSx8egRzz3/PIPBkCiOGI3H7Pb3SNOEtc01JJ7+nS4Cwc7WLs5arLAM+0MasWe3u02SzNPvlXSH4wMqigtFzgEL9fCoV8ylg+8sWhHCzaTgwpPniK3AGk/PFezu77P7YJ1xVkCjRRJLnBuS5RIvY3AeY3OEsgih0EhKSgQSEo0QEaic0oyIbR1HjkESq4Tpdp1Go4FSAlklm4dl0hMYZfIjV+3DffHPGXxHx8TSIzTBQ6PxoFbzgLQ4wsorvQWhyelz48Yb6DJhbrbNH3z5be6vGT71/Cz3bz/k5JmIpLaElB7nc5SOcQ6ElUR6xJlTi+wOdti9t0Ws23hRQ1nLU6cXOH98hvUHW7z+F98miqbwGzv82b/7Or/xH66iU40vYmJvUKak8A7cPs8893lUXMcLR6wEp849yfJqqFtqdY2LGnhvGXa38TbjxLGXQaTge0ROhLctDBPv4GvX38e7ksWFU5TZAO9AeYXwI4yL8N6xuLTE2sZ93njzHdqJ4uyxY3RHBbqWcOHJ8yTliN1HD5Dk3L69hiXi9PEZPH2UqCGdpl2b4tHOBmm0hPcDosjgDTircTZcdZ1yyEiiCTWM9wYnYXq6g3E5hbPEWiH9Tz+gevTMPdgTfATgdfS+sijJxmPq9TplWR4AO4dy5LDeGWMp8wKh5AEJoSxLijJIoqM4puYcLz79LFJKkihBVPWDlJJIR2HPVBqycc7e7g5L8wsopag36iChyC2bmw8pVIEzMNVocfbEKd4b3mUkRhybnsJ5y57p88H1a6zOL7LYmMVlDmc9ToIQDkMAPO0RRR9uQlypLFMmCjwX1iBTeHxlH+adx/rQqDPOktsyeB9mJWVe4IXAYg8IHXGScOzYMaZnEr729l/wO2/8EUNZoHKJQGIlJDbhTOsk/8kX/2e8fO75EBAIRJNjJT583H4Sx/fsO39MAJ+AYL4omWTZMOx2acQR0luMz6m1myw+cZLulbtsXb1NlMR0Lp1i8bVn2atdw15bY+c7lzHeEZ9Zof3cOaZOrDK+/Yjh2ha9jW2y9XVMLPHHZrlw8SJJEoea+gggPzlAk/o5H48RVUNHWYF2lcVWxdQLlnRgyxKJwFlDlCYhNLgKOpZVPVFtZ/ECGq0mtvJ0P9r5/EmeHz+uIavl1wJW+NCUsJ7e3l7ASVINXjLc3KVc34L1NaJeF9HPyHebsDRL0mmTlRn1qTYuCvvxtNYA5ZGZpRwWXLtyjdNPnwXnOXPiOHGS0ttOiI1j/YNbLK0uE7c1tm9Rs/OUg21cqcnLBuVUh1ajzbA3ZGQsc7Mz1OIIYw3RqSWe+4d/l3f/h4gOEfPHVlDtGk1VI7aC48+8ym+tnOaP/8n/m81vfROBJE7S4J0qPQPlqWee2n7Brb94izuvv8Wo6CNGI1zh0WkH7wVxb0Q6sGTNiFEkaTrPfC6wV26xXVrawtB44hJlOoOTEQqHU0EhGkmBtBLngm+qqJQftsL78qLAOEstTamldbwfhnwaGzA7pZKqvrFI6TDe4uwhKOq9xKYR7eef5O7bV4j7BpcXFIuCzAriUoHxGIK3tRAhj8PhuXbzBrOzc5w8dhxvA+tVqkDMENUybyeTZcIQ/yEs/x8LRWi122gdUObAUg2+H1JJdBRVTL9GAE2kDrIzKcGLQ8letQghPM4J7m6soZSixDIaDqi3G1jvKY2lnibMz7SRMjDJHGCKEiU9ubDcfrDH1n4fQTCpH456ECV4G7wnnXWBiYFAeEOdkrlmTD93RI0Gw2GX0sHi7DQn5qa5c+M+/UdrRMdXSYQlrUd4J1G5hFIiytA5JAJlNIV1bPV6JDWJ046ypri5t8PtOw8AERis1mDyjEwKZC3BO0tR5gwHBVsP7xOZgv7mFlYZlk6do3jiPE8utalrh5OeTE+oz0f8DSvZv3chIc07ga/STz8s9+fwMz8ic/+Q9H+yABwpjh4PnYJK5uJCQNZR4PUoC3byPIfPMQFUf7wQglYKL0JCOyKYwgvnw1ysGL2q6gBOunxJFIVuiZIV8zJ0NZVUCCkO5EeTz0JJSRzFB4wmrR2xbuLcbfJszNqjNVrK8+DRI5pTU+A9ZZGhCP6stTiwWssy49qdqyzPT7N07ATD0jM9N0eSNAPrreqqpLUazXYLX46x1tJMEiIUSU1RWEeeF+AtzhrSeh2dpDQbDU6fOs37H7yP97ZiQFUsAesOZA+giOOYTqeDUoruYMA4yxkUBV5o4rSOkpKiiIiikkhrvJhsFMUBezr4/DrwFdAqwv1SSCIdwlWMMaRpilL6YJ7U6il5YZhVs0SxYPXYMrVaHPhoOuaJ8xdZWV7h6rUr3L59Ey08SZyEY1KWWFtWkl11II+JosOUaqlCWEYsg/eWr47hhOE9kdtElSn8T884Elo3uUuEBUNVVhYSTzYcgnNBUiccWityFeTsXlh8lSYuK/uRsTV4G2wV0jRBCcEoK0JnVEbsbu8gCT5ipihwzvPe++9z4vgq8wsLNFpNzp4/R5RoOtMdBv0eWkm6u12UB9lqML26zPTSAtnvfZkrtx8hagkuici9Q1uDyUu63R4+z5lPWzRrNaITKzRaTUYCoipMzgvInGXm9HE6xxbpbu+wee8h62uPGG7t8MQLz5CuzDEzPcO9dy6HgKqtPQotUTMtptptzpw9ydf/IiUbjJl4wNFqIwWYIidK6uTGUavVKPMCVYuRTLqg1Yd+5Lufj487JBOzWTGpH5wH5WjUE+r1mKHJcM5gTEHsYpqNFnEUpD2729voOA1MTivY6/bp9Qe0W21qjQZn2m3uP3hAnCQMx2NKUzLOxhRFTl7kRFJV166Y8XAUbC2mO5S2REjB8vIy1rW4f+8RvcGISSSfmzTwJsyUx6qzo/J5T5B3Cu9RFeiuFKRpwoxrYeJd4kRjpEPLoARQUYrUMZLKXkcovE+wVgNjrLMUuSOWEuGg1ejQqNdwAuoNSNMaXgYw65BBc/R1fT97ir/hVKkfxQhk/zC8PIJScQQAUSjvEJRYr3h4/Tpvff1fEdccN671+eMvv8XQWrJ8EzuG7o5ARFs0pxeIk3ms00hlMC5cm42BVnOeUW8f7zOECiFkhc2JpeT0sRnmf+M19na3cdPzLC5d4N7dW5w7vURDLoSNTpIQIXBSkgiFY4B0oE0LHyXENY+IQ2Sv9BHOj3m4dp00mSWO6yCKSuJnwpv1AKHOLE2OlCBlhLOTDYxAeIcTjlI6+plhdfUM+zv7nFjK2NrYoywcSIfu3aHVbjIYWL773WscO77M6VNzKFViTQIYkrRGv5+TJA3KcfDwNKaH8xalY6QIrBmpNFJESAH4AikVUTLFE+dfwJgQnBk1aj8TTOy/6i16qBorh2CYMYbROKNWrx/UuxM1UGD8hHqqLAqcdehIU5Qlo9GI0XgcmuoTn2YpaTYaUEk/JwGAqlI4WmuxPriMxnHE/Pw8SEFar6GTiF6/x90HjxgPesw9scS79y9T70On0ea1mUucnT4JssVWkfH+9cssTs+z2JrGGxtcoOVEjg9WhveKr7y5ncdNAlFKcwCY4h3WGbwXlC5YqWR5zjjPMdXrz7KM/mBElmfkRYatPF998A1DAc9euMiTp05yfe0q/98v/wse5D1sWSLwWGeIfcx8PMVvfPZLnF08hfLBxuVQCv4DHMgfw3jcOu6jfvZR+9Gj44cFtIrwRwN5Ah/2eN5z785tfGlYXlpkZ28XX28yc3oV7QWj62tsfPcqKk3oXDjJyitPM2622bl2i603r5Du9nBnjzM1M8v88xdZePYCZWWHVkqYW12iPTONk+Jg7334YghKOBFIOd1+P+AXzpGqQAKpVfkPSqoA2lfqG+tsaLD6AArhQUt1YFsmqOTUStNM66gJACR+XmMeHb767GRFpPICcmtY39oKxCnnKWzOxs4jurtbzBQldPvEY4vJcjI7RgyaiFYTX5aYRpONXp9jx1bJR2MunTmL8IpkqoUHrr7/AYmIeObZ5zl+8iRaajrNNvW0xo1bV4mjmOWpWdQgwzcTjLAMhMWWhrnmDM7D5vghs7PTJM060sP5S09z4b84BsMMudSh0BGxjHA6KGoXl5d47pln+fOvvx4IU80U5zy69BTekVPiEoXbH+BMQRkFix8tIvaMxSjPsgiEvJm0hdJ1eoMB8/WEYq8L717DZY7puElxqcY4rVPLNJGwlcImKDzcwd7fHuz7Qy/KIZxgPB5TS2t4YNDvB0WpCxaBeFAyqKciHVF4g6+k/845YidZOfsE/V/8NPtbu+hOi9J5EieROKwOhClnbAiSrLoNeVHw3vsfsDA7R5rEldJtEkp35Pr0Q170PxaYOjU1BQS5/wQQgcPuk1KKNEnw9Qa6ulD4isILvpKgVyo4JXFe0B0PApgq5KHPTJUSryNFmqTIKCIY9yuMALyhzC13euuU1qFlhLOW3IyRWhJHHfCKmqoTSY2sJxghMR7y0jI1O0NWjokHDYxxICKWj5/m5rUNbvfGPO0gLi1zRIytZ4zASI/UBo+lcmfEKtgb90n2PNKBaQlev3mHwWAEtTrOFZhiRBxLhHXYwmBUgVKwubXJcGeDfH8H6XLyQZdr6zv01nfpP3WGT1w6TTMGoyxWCKQT+CrIomqdMrl0Ch/MgSfgiXsMMP0oCf7joVMf9bjHmafW2Oo1cFBMgfhQ4NRR8HUCpk4S235cQ8iwkXXW4aU/6JoHFqnD5FkIdVISrQOYarQ88POYUNDLUoXiWwjKMhzj4BP1uNScEEjWmMJ4ybC3x+X33kMVOXlecGLlFGY8xvsS5y2R1kRonnv6EhuP7iGVo96oU1iHVDHNepOisJPKNjBKRfDv00lEb3+faJTTnpphmBW0mu0gx8azubkZfEbilOmZkJA61enQ6+1jJ4zkqkA2lZwUHwLkhAjntq/e32A4ZpyPAy1feJJYE8c1lEyq4BIRpHI6wStXXawqA4BqGrmKLa1EYPuOsgzrHM1IEanAMkvTmCTWSJGwvDRLkiYQLtMgQh5vuzXFC8+/zMrSCjeuXSHPxjhrKU1ZfTbRwTw0xhzM4wm4avP8oHEwAdDDJiDMY62jKlLwp2CIwy+O7vEnlw6tApCKNWRZjq+aTl5FdPd3abYaZEXJ5to6p86cZjAc8ujuA44fW0XHMePBkKXFRUoX5pyqzMkzb2g1mtTTGsPRiOFggCgMEZKVxSV2d/YoSsODjUfs7e+xsDBHu1VDOE85LrAO6jPTLJ0+ycoTZ7jz5nv86f/wr9BaQi2h8J6iMGTjkv5+H4BWrU4rraGloh9JlHfERCGEp1pvEOCUQEYJ06uLLCwusn7/IaM8482b13g+ibj25nssnD3Bw91tHt29z4NHa7z2y78EacTxY4vUGwllllMWASzL8zyY/7ca2Cgl2+ujhMBbi/SiChMUh5zEDx2Iv5njxyHX+/DfEIFlGXxZELiwWfKWOAbnM2qNBGcstTRhOBgEoyxEFdogibUiz3KEUHij0WmCNYp7d9eYm59jOCpAJ2SFIcsKalEN7x31Zg3nPa1mG60jNt0mWTZinOU0GnVq9ZT9bol1BUIqer0RzguUAD4EpH//EdLPBUZ5jAbtQXtPHGvKWGBtCDuMNNTS4Gms4xpGSFASqQVC6BAoRVAgWBGjHWgFToNC4m1KRA62BOFCs8qF5t4PNjd/BhCnH3B8j7+sDzxUhA633leJuBJEnY1Hd3j/67/H+fkWmZd89TvfYHUq5dKZ81x64TQrZ08iPFi/xdajt1heeoJG6zilhwyHEzGR87jCUW/NsLO3TqJi0voUohyRlAKbj6mLkr6O+NL/4v/AwvJpfuf/979ntdtAtaYokxzvLGUhyCjp9R8xPdVByhrYHCc8XgW2h0QhnKMc7/Dw0XU++YnfRMgI7/pABKKsgBMNNvieWmPY3d3Bu2o9rph8WEtpYZCP2dzb5uoHl3lw+xGJjnClpdWM0VIw6Pd5/95Vrt58yCde/iSrx6eQKkOJFGMk9bpkY2OdJGrSaE4hIhDEOBdTbwisExRlUQVNKUAjpMGJHCk7XHjq07RaJxjkBSYv8LU0SGl+xoY/+OcPQIeJLz2EGrderx/UymV56PV4yCDySKlQUmFcABgHwyG3bt9mY2uTT736Ko16I1grHSFtcKSJXVaNbj95DMEHupbUSGoJXsLdtfsUwzHPPfUsl3fvI1XC3FwbuZ/zzPQTRLrDrh3z3rW3mevMsNiZRZaOwjgcQRYqXGXX5T2SYA+FCOsKMnhMW+cwZYkpywAeV69V6ohinDEYjsjynMFwSJblIWRLRCgdE9UkSobAROEDYePC+dO88vRF7t6/xv/n9/5r3t28ycAbUueRzqGEpO40f+vJV3n5zIukIj3YY03q1Z8kGffHrQUeZ6nCDw9APfJHqJ44qF6FAO8Y9PrcuHyVWGvOXDzHcDhCGEuGY/qJU8y0p9l5/yaPXn8H1x2x+NRZpl6+SG15ho23r+GurbF3Z4t8YRYx00JPNYiSiNbMFMsnj9Go17GTkEkxaaB9+L2paps+Go0wWFSUUEjw0pEQLMgi54lVCKOK0+D17qsmZxwFhZ8zFoQMRBjhUFHE9NR0sIETAi8nORaBCfuTMUv+ekfoYwaLRic8TsCgyNjpd9FxRGQFo/099rfXiZ3F9vo0vAwq4rIkHhpc0ScapMj9gvrsHGZzHSMlt+/d4vj5c6TNKU6dOY3Q0JmexY8KyqLkj7/6p3zmE69x/ORp0npMe6eJjDWbDx9x9833+cQXfxGRF3hgNMzJjUDLaaJ6wvb2FjXXYb7WDmXZ0lywClGCRIZ8lAKL8BrnchJTslprsbXfBQSFKfClwwlJIQMGIkqLdh7hJINGRNSeoq6aiE5Mb9wl3+9j+kNi4VEC1na3aXmJ2NljaK4xbtSZmW2iV5Yxsk7NBNypALxUeOmgIvRVhjkHQVLGmCqM2pDECbIt6ff7FQmikvxXthdKqKCUmmS0OIvLYeQ9pz//afJxjktqaKdI8dhKJa69pRQhqwWqa4yQ7O7uce36dZ568hJSwaF6qGqA8MPfln08ZmqrBUC9XidJkg9d5IGDlGyJOAjP8SIsVN4Hv8qJjMIJj7G+8i6LMZGlNB5jQqfuQHYmJFppnBTEKqFerzEyGQ+3N1kfBxbfKMsYZSMipcGCFoKo2aBIIlykaSYpkYoREtKpNg+3NlleXqHemqbb26IY5azMzWFjyVZ3CzfuE5ceFc+QtyS602Gx2WT73i3KzQ0iF1GYkJY78oa9vicmYWNkuLqxibMQSYWxGbGLsUWBUAlJHDZ6o9GA7fU1+lvryCIjataoTXUoBobNG1f48tptnP47fPrZSzTMgFyGJDmc5WhCdJDqTyR68vAz5sNA6MFjHwuUmoCHj4OqRzvPRy0AwmM+DKa6x0DbyeMmYKpz/scOppZlKIasdUhpKYoiAJhaoyONzxxSgtYSIYME59Dvs0o+rsBFWflsTDrrE49gWUn/hYDSWLSK0LUWFoHLBpw9c4p6q8nSUovSg3AWLwxKy6oJoWnWUwYuZ2FxCZSicJ5ao155UEG/t0+jnlQM71DIjYdDjIXxcESt3iJNY4pijMcTxyknT51iPBrx7W+/weDtAU8/+xTT09OMhoPApAO8CCnQIRQNrCnwZXUOKxWSV2v1EOAihoyzgvFwQBTHpGkDrVKkijDGhgK6ukghAqPA+cNkaSl8BaofzrkszxFDRT2tEac62ILgabfqpGnMhDAwqbmUEHgUUkuWl1Zp1utcu3aZ7t4ORZERTYLu/KHv76RgU1VTpygKsvE4sASqx048WL33SIIn89/c4R/7arJMVNcKKZBCoQRI5zBlQZHllEVOmeW0W020UtTrTWIdA5ql2QWkl9STGscWV2gkdQwe7SRbDzZoL0yz39unrkNIT2ZN6DoWJjBAjOGbX/kqs/OLLB8/Tjw7T7PTZpwPmJptoaTAG0eeG/rDDN2e4sIrL1GbanPn7Sv81//n/xIrBWWnTukNtjsiG2dY50mTlFqUUOQjdvv75NbQaLWhtCgvccIhIxW6lR5Mf4RUoYNqkpjOyWXmj6+ytbtDb7/H8qkTbO/v09/a4/Nf/AK7f/AH3H3/KqdfeJIiG5JlQ4wtK9m5IM8KxnmOJ3hs6khTWkucJIds/2q+HxDWvpek+PPxMYYgMOCF9whC97nTqhNHgq2dLvV6C6kVSZpijCNNaxUYENYjHYWwpX53SKveYTDokcaabDRgqt2kdCXNeooUjpoKioHClFWjKA4qnMryp8xGpNEcRZFz9ux53n73NlEc82h9iyK3pKmq2J5/GSvn8fLtCAvUgRc+JJMKQeKCmX/hHCjwWlPoIJVVUoR6R3qEd0gMQji0cAgUTtrq+mYRwuLReGXxPvj54itARFQb3o85MyvC7ZH3Aj/8cvRv4PCT/45+Jj74SwsQ1gMlzo3Z37/L/etf4eyqJHURSaPDsf/g73D57W/z3Q/WoHGKdx8UrN19yKsvH2O/t0e/f5dTJ2LmFucpxz2k8ESRpvAGazRzzRlKWzDc36SZNolkg5HPyL1n9eynWTr9LDtbV1k8/hLvvvU6L3+mxsjHJLZg7CWjwT614RILqo4SMQiDco5SljhiZCnwdOnuPmJ2dgEVpZXqwYEroJpjkw56XmSYomSqM3XApLKVKsYbS2EMg/429+8/4P7aLpdvbdAbjCiKDGuLQLjwkrnpGq+8+jyLSzWEtEif4qxDKoeOasRJnanpKRQRcWx5tL7LyuociCEuL6tAzAilAkDrvEMlTU6deZljp5/B2BA2m2VjlJrmZ3ou++DmbIxhfWODwWAY2FTLy8RRfFCTyapRNckggCrsVamgBiPsBZ1z3L57JwTkeEekAwHBVnkMUqkAHBEYe44AhiHCvkHJ4D8Z6VCbG29BwoZdp9ars9iaZVqfpsgykvoUY2Poxj3euvomi1NNVjoLYAXOeWTl1xsk/UGCGmtFpBMipQ/S0Y2xFEVJkeUVW39CCAjX83Ges7m9zd7+PoPh6CDoMHjyKzwOKRQWAVLTSBo8d+kSJ04vcOXBe/zzf/PPeP3+25iyAGsxHhJCs+vMsbP87Ve/QFPUEb4KvT0Cpv4kju8H8P5VmRk/kvT5inksgCvvvMe9m7ex1rCwusTc/Byj/ghhIIs8nbMrrLQa7L99g+GVu9zf7TL97DkaJxc5MfsJyhuP6N/doL+1R7m3TyE8aqbBs7/wKRq1Ok6KA/LIZB2Vj70XKQW9/T4z7Sk+++qnEFqRD8cMRiMyY4JtnJRkCmSkKIVHCx8CkA/UpzKQxHQIzxRSIiNNrdPCacFBiq7/iKbez/AQ1achfCX5B3q9Pr4C7ygcw7Ud6kNH6jWZNWQU9HVBggPriUpHkuWInkL2x1yqp6j9Lot4GlnG2r3LRLUWKxfOsDSzQH0pxSI4s3AMhODevQfk/R6nnz6JM5acmNoTF6gbwdtf+xoXXn0JOdOiK3dJFNTkFDKNGXb7qNzTancohMBribKeWAaVpQDSVOMGBXv37lCzBTP1lP3SUhiH0GGPmzpFbD270iFwTMUJp196hnR+hXnfYe7sCropePAX32H83ffxe10i78jJEVqCKtHlEPHBZcTSHDOvKXbmjhOjcEQU0lMKjzMGpcKaHQkV1nrpq714wFDGeU6KII5TGg3oD/q4ojwgXUY6wlp3QIAK+IQjcwKpY7TyqGaC8RHWQSkEQgm0Ddd3oojC5AFQ9SE8sLQlV65fZ3p2muWVxWDf5kM98qM6VT4WmDozMxNYb1FUganhQwqgUpDWRkrj4yQYbgdaX7jgVB5y1loEIbFMeMni/DxJmmCMo7A5xlioupJaa4RzeFOiJmlnOEoZ8979DUoR0Wk1mJpuML/0LDKqY4WmQLA5GCJiDVFMEqVI4VFaMj3VwekImdaJ0jb53hYfXL/D3PI8ql2nyB39Yca42SQ5tcxevs+NR2v03nuXs82UulaoKCHPS5w3we/BwcB5rq9vMMwL6lEN5wy+yIOfZFEiIocxHl19FkEvb8h7PcbDMaKRBC83VVKWEXd3Mp40ES00WgoMRzxP/SFANZEVTu6bgKOTxx0+3n8IHP3L2KtHfz4xUZ+AU65ie05+BmCd/ZD+5OjvTVipP24w1TlHUYT07YlvU8VXwjofzLwJ6aF4d+Dre8BSLSu2YxS8gL9HwlIVgmVZAcjeoZVG6QTvPZ1GyrmzZ0jSGmZckJUFmCIwTKVCqhglIxo1T62mSGOFMaC9wnuJFJ697j5raw9ZXl5AiOD/muU59XqD6alZstGAve0NWlNTWCkQSoFIkFrRmZrmmWef5R//43/MsWMrdE6eIIkTirI61tIFtonQKAlOGqypZFtlABQ9kCZJsPEQA8Z5wWDQp5Y2UCpG6YiyNNWGXSNk9Vv+cH5MwHoPIMPcrNfrGGvJ8wKJoNVuU5YFQnparVoFvk6Wwcn8JBw9AUpFdDrTXLhwicsfvENR5vhqjk3+FUVBVNk2TObexKclSpIDwFUpdThXyhKo/cjn5o92TPwYZfW9qBQPYWGRUoI12DzHliW2yEkiTaonm+DgAwyQxJpoJviXRTKiNj+LE55YCBqVlUvhQuDJWIyYmp5mnGc00hoiVQy6ffY2t4i1ZtzvcuvakKUTJxGxot/rMdVpYI0hLwx7wxFzK8d44tlnqTca7G/v8Tv/3T9n6A1M18nzjPFwAEWJ1hGz83OUw5yi18dLR5Qm7O/t0Wi3wsbDWVIjiWoRo6Ikrac82tnGFkUAxqxFak2kFPMzc9j5eZRWtO6vcXOryyAbM3dyFbojkjhmqtPBmJKiLENIj1DoKGJclFhP2Kh5yyAbB8WFDMAf/iNW66N4y8/HDzDEQZNVCU2z0WR2psPDrRHGuMDMlBahNPVGjSzPycYjsixDKU2R5xxfnacsDdOdFr1+L1j/1BLGOQyGA3COuCaxpsTYEC7SbAXW8+xUC18OSLRACUOej7hz+xF5ntNqt9naWqM0jhp6wu86ijoevIfD8b2AqvQBgHdS4oC7d+7gjEMh0d7jrUVoHRgq3h6wHZUI4IYXAkOMQuG9xUuNJ0W6Ai9LjCyxwiIIzTwOuCyCH2RSiscYOD8GwvLfwDFRETlA4Z3nwZ2vMdi4QnfvCqlKiKIpVAKD8QZC1Xn1lSeRaZOvvP0ug/0dlhZPszVq05z/LO9+9zvsFTuczj2nj80gjKPMB1glUWVJ7GvYMiPbe0hrYZHM7hIlEWuDOksnXmLQ3eMPf++3efKZ1/i313+H85fa1DoruKjF3sYNssEujanTFPmIpJkgbRPhxkg3DsCQ85TlHl//6h/zxd/8R4g4APPCehBl5U3qwed4V4ILHvNbGxvg7YHabKIYKoqCMs8Z9DPWHu2w2x0yGI8x1tKsNcCVHDu+wMXzJ0jTEqmDL3CIEi5JE8ndO/c5ffYc1udoAV562jN17j5YZ6qjaKQJSiahplICqTxS1plZuMj5Jz+D9RFFOUYKiy1MePlC/tRdnj8Oi9ATatlRNma/2+X+gwe89dbbjIYjfvVXfplTJ09Rr9Wq3ISg6MnzvFIz1QN5wRhMVW9JpUjrNT73uc+BFNQq9qm1wVlPR8EF1BwhZEysl5yxaCVpN5soIREqACJaKqanpvn9e3/EzY1NPnP8JZb1DPVkjmw0xtQF79x5l06zxfLMKrIQlATBkXA+ZI9JgVOKJE1QzjMejehmGaYsiaqw2TRJabdbONtgOByG+rAo2d/fZ21jg739fZKkRrPRBDgkpWACOUOEPedMZ4annrhAsx3z+gev89//yT/jg60bbJd9KAWpPexMdepTfOGTX2BpegnhK5Ys+kMWaT8prNTHx18m9//LgNTJ5zX5+ofBWJ1IiyeqlK31Da5cuUw+GjMcDrl+9RoXn38WnaSMxmOatYTCW9qr85yenWX76m3Wrt8m+/o7tB8usXrxHLPPPUH74mlGwxG+NEghmD6xQmt1jlKEVTtIhAObWwr5Pb0YK+D++iN6wwEqiWg3WqwuLtFqh2yYMi/IhiPW97YZ9oPFSqRkBQSKKn8mEFUKY4LiVytQklqjjpDhsdIGRiociHN+5kdwyK9w5koZ0d3dRYlgrWCKEaOdAVEmadU76KU5BrtrFFlBw4WQcp97tBVEbgj9DGUSvOmzEEekozG14QhJRLG1z50bNzn93JMQRTy1eop+PSErtmjIiH4+5t4H1zhz6TmWjx8nijXLx1eJleJb3/k2T7/4CkkhwWvcbJM4DdZTubHMzy8SeUVpSiwObx1pHCO8w+72yNceMSUcOk1R0/Os7+4QmxxfBFXGyHnGtiA1hrqIiLtdOidP0mx1iJs1kmMzPPvrv8pdFfPga68Hq4MkImnUKPMRMY5lrfE3bzKcitGvtHBxB0+KwoDIgQLQWKlQXiIquX9Qi06IfIFI5b1ARzH1epO+6VZh0FXwmi0OGnETQltGsKuSPmCDpbCUCByOFIkSAqkV2kmsD4pb5wLeh1BkecG777/P3OIcMlLB2vAop+GHvCf7eMzURiMwc5QKYIVScMDoqoyUI31wsTxgPkAo/KXEWxe6MTiklSwvLlFvNChKT82bCoBzB0CsjiRaebyQeBWYbnuZ4+qjLR7t7HP/fk4qSq7eusvC+eeYnl8gjSRzrQbEdbbWNxBzJa1Oi2bUYbYzT3f0AJRidn6ekcm4UyjWepZzr32Jetrgjb5jP25y/5vfosGQU62UmaUmZd6jTGNKJ4gjiVYCLUDphB1rWe92Q5ErLUrGVSezIInqCKUJPONgeyCkJK01kbUMKTw2d8StOqU1NKZbWOt47/I1ps7PEUcVFCgOpasHgKn7MAP1KMP0cR/UoyDoZBy9/yPDqL5Htv8Y8OqDefDjXqsTu4ajz/fjGs4GWU6apkzCo4TWFGVBYUKKu6wWIWOCN9KB9NuDruwrIh1RqyWVzL2a0xC8TCVEkSLPDRAA2ShKUNJz4tgKWkWUpcWaAmdMaB56CBzIEPDRarZwlJRlhtSaLCup1yHPx8F4f2nhQO4UgpJSGs0WaZKQKE+xPeDq+2/TmZtl5eRJhATjg2ypMzPLpz/zWXZ2d1ldXQ2LrhBIqdC6otBbDlIjg7exPwKGg1LhM2g1myCGjMqSbreLUhqtQz/dVAxtOQEQlDhgNh8WTOrgviiKqNVqjMdjiiJ0rYzxqCiuAqCOViMBnA2drYqtWiGz7Xabs+fPMxj2A1v3SNrl0Xk7me/D4RAdRfiiCLYiZXnAnhVCIKRiNMp+1FPzRzYmgDMQrCZdYFMH6URgUnvrKEZjhCnDWmIt9x8+wBjD6ZOnkEJijWX90SOsNeTOsLqySi0KfjNa6xAciMMpj81zjq2ugoAyL7lx5RpT09NMTU8x3u/yja/+BZ/9zGe4f+smW1vr9LMx5596kpmpaXxZMOqN6OWGs08/x8rJU4hIsj8a8Ie//S+4e+c2Yw3F/h7lOCNCsHTyGNJLttc3yQcjlBA88+rz/O1f/xW63T20FNQbTQwWJ2NGvT5KKwQwvTCPUIJsMKIYDElQ+KbGR0GiWIwyWjPTtJbnKY3h4pNP4kvD7vYWxmZ84tVP8vDhJg/v3mMwGLHf61M6G0xf4oR6FDF7bBEnq2L24Hgczmfxc9DpBxyTcnjCnAwKDLyiWatxfGWey9fWcUYw6I9otdoUpqT0JbnJQFcWQ16i45jdvT2azRbOQ63eROkI5yVF4Wg0pugP+kT1lPX795BKUavXGWUDptod8nJEvZEy3WnSatYoywHtVod7D/fpdObY2s4oTTDVn7z2v3qEx0kP0kmMDKzUcF9gkf/+l7/MuScuML+4iLKG2HgSDZQWFyV4EhQaKT1eDBHKUjoQfkwxzrFC4UtHXYXQFJkcx+saXpaoaiMofkAg9efjLx8fhsfDmuUIAYmmGHHtvX/Dsh8zl4DTi6ioiVA1hCtJ5Ji9vufeg5yH167QG+a88a37rO8O2dvv8sonXuLiKy8y2HnE2r6j00hInQphDeUYohyTJGRFnX/5u1/h0oVphJC0T3+JdnOa7dtv8sUv/j16+S5/73/7X9BSI8aP3qDoj3l0cxtjBpx6MuPWzQ+48NwvIRV4mSBNG+cGOEZcf/ctnn/hZWLdwPkxwmmEAyEs+CochbyqoWBlaYl6GkO1ARQHChUwpaW/X7C1uc5+d4uiyIlUnVoCruyzMFvnxGKbSIeNFk7g5RikoJ60uHrlLmfOnQuEBhWst2TkaUxpGs1TCJ+RD/eo1VpYDE5k6LhDq3OS5577JRw1snJElg9JdPUaD47gT8/58P2AVO9DQ2WcZXS7XfZ7Pa5fv04cx3zpS1/ClIZmvV41xQ+VP8ba4M9bluR5kLkfWCtpjQAajQZprRYUGcZUDeywb7TOYZw92FMcvBZnibUO6scork4qi7dhc768tMTn3Wf5o/f/gq+O3uCLz/8CtTFEieD9W++RpjHn508RZZpShJAUZADIpRQhOT3VZMMx2WCIFy6oEAgkiTzLGPQGSKmo1WrU63WU1uz21rl16yYOwfTUVFUvAwgiFR7baic0Ww3qjQ5zU3M005jdwRa//ZXf5V+9/u94ONpkLx+Euts7Ui+JHEQ65dNPf4pPnn+BhAiUC2Qcr3407M0f8XicxPP9xv+UsKqjz/uhTACgLAvefecdtra3MMawtLhIMSro7uxT77Qg0sHiLCsooph4rsn87LPMnDvB+jtXMbc2WHu4R/f0CsnCLFEUM7I58+dO0V5ZqCqRsH76Iwy3j1R4WIcyDlUYmkmdltc0hEaXjpl6E9XSJIsRT8WXyItgG9Hv9xnnOUVZkOU5xppgx1aF9mA9zShloTODPhSqfmTv/md9OBFY88ITLB66A6T1COfZ39tnv9cl8VBEinRlCdeO6G9sUez2aCNIkhhrShIyhPSIckQ0jJHjBOUFi7WEIhKU27u4jR2Gwy47owHpvT06n36WMxfP0snhSv6QzZ0tzmvNG2+9STLT5PzF84g0YiU/xrSKyW48oC40I2+YWl7ACk9ZFHS3dlmcnUfoiKzISbQOa6lwsNXDbe9Tc4axlLz2d3+TD+7c5uEf/ynNsWHdZDz0JUIo6gJSDEv3NlHpbYYvtWjoGcgMZavFyi9+kp2dLbL3rtIvxjD01LxF4Gl06rgIdq5cIZ1ZIH/yOYyuIUtPKmFMCUJhCUpdeWTvD4dqZ1NanIUkSUiTFFqO0XhEWcn9nQ/71iiKDvbukZO40gXVFh6vAxYmDKFRqxRWhfwPHWmMs9iJiloqvHcMxoENHk9FYU8gZCC88MOX+n8sMDVN0wNp7IFPpnOU1h74TEKQz+oqUR0RZNSTxzsZPOWcE2ipaTWazEzP0B9k1JstTDZGFCH4h0oaJIVAC4WTEh8lfPf6ZYbGMjvVots1KEq8lAyHY+J6xsLxJZRyWKERosX7b34HZXOm2k2mp6ZZnJphLyuYP3EcfWYFCZjhCE+OtJaSgrFMuPTcp5gXjrTYQ3TX6dkxAoV2HlWPqCkZbCJUwv1ej36eEYsIH6mQDi40QkUk9QYqigLyXga5sUPSnFlgLCTO5dTiBJwCKxhsrTMcf517aYoYXuIzLz4dQjUqb7XJouNdkPh/KFCKw8Xpo/xQPwponYzHF8GjICwEgGYi0Tlgt3IIWB0yUg/B1ENv1R/fKEuDlvrghNSVhCe83xCwo6RES0UhIYr0wecVvPbCCaZV5aEqDq0AtNYHDMgkiioWpEbpmDhKmGnVOXfmTLAawONNCcZOiH+AwFbdTKytNtCaLDfce/CIVrNFf28LrRW1ND343IvS0Gi2qLeaoXMf6UqqDv39PepPnAu+JB6MdUgdsXLsOHduXqesFmEp9IEnmJIKpTyRD0FNzk8UeuG1GeMCo9eDUpp6o4UbZ4xGgwCeSkWt3sbaGGtDUIyMRMVacAfzK9zKD8+liqHq3IiitBSlJU3iwK7CfejK5g8Y2YdtpACAKmam5zh27BQ3rl6mKIKFwsTL1hPYqBBkWr66dVUwgrUusCO8x1pPWeY/dgb1j2JMpEG66uopUaXIm4J8lFHmOaK6pjpjaDWbWGN57913MWXJqROnmJmeRitJb2+ffHuXshYztiWbd+4zGgx56rVXiLWigaI/HoNSCAtPnD6HSCJG2Zj29BS/+IufY23jEVESs7yyzM5wyDgbIYVk0B/jZcL5S5dYOHUaIyVmNOL3f+9f8+d/9mdE45JRt8fUdIepxWWaUx3u3r3LuNsnLhz1KCZZ7PAf/e/+M2ZmOkSR5uG9+7QbHaamphlofyDfM0UJxiHR1NIazXqTYa+PiyTjfIyOYyhLPIKzT1/CS8EIi9SCTtKgkDGfePWTvPHWO+Acg/6Q3mBIZnK8IIRsqHA9OWhzPlbgh/AVz+Hy/dM7Pq6f2vffOHlC3mbwCw8cP0IopPZceuIMX//GFXSjw3A8JstGDAZ9hsMecRyhdYxxjqn2DOMswwqwwgUWv5JkRY5XgnE2Ik4ihIC93pB6a4pOu0Wv2yXPClzdsbfTZX5umlq9zf5+j72deywurtKogIa8cOzvD+m009A09pPr1ke9R/Ghd+iEwAmPdsHfyyhBfzhib2ePP/+TP+fXfuu3kEmM1jG2lDgMu1sPiHDsbG8wzgcM9jYYd7dYXYmZ62yT97fxOqGeFsTUUfVPEjWe59jJ8+hWG3ew8/rpnos//lGd/wf+9uFrm++j7SZatTFKoFROImvoOKYuFVk25Mr9Db751gdoUga55x/+o7/HtetX+cwnP0mv3+Of/re/z2c/9ywXXjrHoGe4dv0m2ciy3K7RqHl8nHH81Dx3b5/iq996j+nlGf6jLz3De29+DTfcQiR7dDoRL7zwGg8ePGT39lu40SP+5JvXKMY95s6c55UvvoQ3BT5K8UJWjI8x9268iy0yVo6/gLMWpQGXAQqcqkAEg8AFkF541tfX2dnZ5MnnX64+muCBZp2jtCXbmxtsru+QjUqK0pA2FIPBkEhKkrQJCqQoApPah/pZYkhqNVaOLZPURPC9FDWMlURxSqItjx7skyrAmCqI0iEjQWtqlede+CVk3CQzGcNxn1rcwNuqRvAGIaK/ronzIx8H+4YjxAvnHFmesfboEQ8ePGDt0SPOnzvH0888EzyYhQzHW1VEEAIRRBGmthkH5t/W1hZlUXD27Nmg3NL6sT0H6CjUeJPN7ofZ7dX10ntMWbKxvk6z0aBRq5MkMV54pFNo7/nk3Itc/PQTXNm+yaO7Dzhz7CQ3H66TxnWeWD4DeUiWLpxFEprIQmrSWg3hHdlgiHCe0WjEfn8/5HNUVlFKaur1BkmS4KxlZ2cHrTVLi/PMzLzGYDCk1xtQmhIpJK1Wi06nQ6fdpl6L0EqhdcwoG/Hu3Xf4H//8X/Cd+++ylXcZ5zm2CCG+SIVREFnNxaXz/OqLn6fta0Ep5oq/xCLmJ28cfZ2P3/5l43GbuaP/Pg6g+vhjJn1rUVkIvvmd7/De++/RbLcoekMQguX5Bfo7ewglSes1RKIRCLJxFo5ZvUFteZYzc5+ifLDF9t0H9Ld26W/to5KIqVOrzE5NcyDl96HFG0I1wquYtCY/9PlYx96DR6y/eZkpEVNEMWWzQ22mQ7wwQ312mtrsDGKqQSOt0ey0mZcrOBFUn8IDNuSVmIoNPhoMqacpcdW08HDQxPeAdj9f1SFIuT3BK1VU6h1bGpQMe59xr4/JhrTTEODtkxQvphGlIlYdyt19vMsrlbVDO0ssPMIYlBfYbg9Ri4nwxIlgdbaNdhnF/g6JzcmGPe6ubTEtYtpnOrzy6svoWNMf9NjzYy4VZ7h/6y7zp5aJvaRf5vR6++wPt2nWYlSjDlFEVozYHuzR6rRRSpKXBapWD3uP/SH5KCePIszCNO2L53nuhad4dPMa8Qd3qWeWUjtSFQKdUzxnhp47b1/nG4NtZhaP87lPfJb4XBM7P814tkXfB4sDX5ZhTxzF3Bju0IlLpnoRoze/izl2DDffpFEKhIUSgfUWnMSjsEdwqaPXCGPLAyWT1kG9YL3DjkaU5lDNCpWdjBSIItiFOhGaUFgXAl4dmArs9ULjvUErSRxHFIXBWIeo9v7D0Zh7Dx7QbrXQSKRzlZ1iAHvED9GD7WOBqcGT89AP0fuqVvSBMaKqJHSpNFoblNJVOroIPpGBJkKQjpf4Km39+PHjPNjYItYRTkr6gxFRPcbhyUtDVMnmSyO4v73JGx9cZzQe47JhkCs7cEWOKwaMszpvv7dF2d8nHwxYmJ/n3LElNtbXuPb+u0gRoZOYs8+9SLqt2NjaRtQaxN7T29xjZ2uNIh8TacWdqE46M8UTCx2Ox2MkA1DBDzVRGmEMVkUMiFjb72I9RFqGsBOd0GlPU5+eRXc6FFbgfInJCorxGGM9UbOD0CkkChsZ/MiDbpBt36Hob6OHKa9//U1qccpLzz5VBfIIQjK7+5C+LixQk3RN8SFQc/Lzo2DnUSbf4e/7D31/9LmBysPn8L7gQeS+53cP2YEfvv/HOZRUlZ+nQgpBWcn+I6WC1MdYdBQTxxFKa2wl9548PgCo6tDeoJK/ax2k/1mWHXgEew+1Wos0jvn0Ky8wNzeLtR7jyyCbNwbrg6TH+YmZfbAaCNhixKDf4/atG6wuzyNcRqczRV6Gbn5RCNJanXa7SXC1FwyygkZ7ilWtGI6GFFlBq9bG2bCiSiU4ceIEiwvzJHGEd47trc0qqIzqnFQH8pjD4x2YqgiHsSFxz3mP1hHNusb7HoNhH+ssszM2eHCKBmiNlAonJ+8p+PlKqFjlvmJWe8qyIEmCpCrLCoaDcfCZ8oEdIL2sOluHHeujtyBwViBFwtLiMe7fu0O/3wcnMdZiHQfS/7IMniwixPxRFCXj6tgZF+wZAmPeMhqNfkyz84c/jnb3Q8MryOKE94z7fZwp8dZhsjGmLGm1WtTSlDjWWGNo1s8SRzFXr1xlNBwRC8ncsWWE8GzfvM3FZ55i8eV5Nh+u4XoDXJpg0xo4T1FkpDphenqGYT4mG2dcu3WbJy9c4ESzxmg/dIBPLMyD1uzt91BJjaeefYG40cAKgbeGt7/5Hb7xb/+YXrdHZBznzpxFNVMGwxGX3/sgdFCVJ401M7Oz/MZ/+h+ysrjAN/7kT/ilL36B80+cZzTIyMuC0VafRqeJq3kiFHiPrOaESCSqChiqKY31DllP0F5QZBllViITjXOeuF6jzBwnT59ham6BP+MPuXb1Gi3dpLA5w3yIEjVsVvLozh2m5maImg2gagxUjB4RtEb8vMz9AcbkvD9gT3qs8wgR0sKfe/ocqTZYlxFJR5ENwRm89YyHeQiV85I9IQENXiNcTDEOa1+72cZaSywU0nrIDaO8B96SesNipw3OoUxGTcPe1jqznTqNRsTC+bNsbnTDq3KhIXP//iNOn5qvdPAfbzPsBRjhiZ1Heo+VAXjPbcFmW/LNezc4dfsGT595IjBhdnfZe3idh9ff5fj5s7zb2+XGaEz35hr6/ibNVs4vvFbj3LECqUYou0FezFPIabb2ttgcdPnMp75Q/XH46Pn4YY7lz8fHGdXnJI40AwUEfxvP7uYtYi8YlBm5b9ByG9TauxS2Q4M2+0PBN967Qbc04BO+8MuvETdgYXkZlMcJwzPPvMh777yHjnq88917zB5b4btf+wvOrBxj9dgKZ88u04wtT79wmoEumT/3HI3paTr9R3ROv8hgb5uFYycRqsaw3+XEU7/I63/0e1y7vs3pcyucuvg82xs3OTFzkcJ3iBmDMAxHu6yv3eTs6dcQUTN40BkFogAcuBS8AXKsDzWix9Co11DMMez3K5AyMA+tdRhrGGchxMc5SNMa+7u7aAW1NEZLRyRcJc12RLGj057m5vVbDHqW+aU5PBYtYwRx5bseI7VjZ+cmJ5ZWiNMa1hcInaLrbZ5+4ReIa7NkpiDPRtTTBsJHOCx2ElzxU+rDchSA8h7KKlgJoCwMVy5fZX5+nk+++kmOHz8e0pWtqzabhwQDKULYjRIifGYSxnnG2++8zcL8AqdOn0ZpHfYXMtTOQgi0CkE6pTXs7e2hdESz0Tw4VZxz4NwBOLS/v093f584CjY79XqNWqOOqvxTp2WTl048TVGMGPT7zLXnmW8vIUpBqQ3eF1jrUQ6SOEbGmtwWSBcY0vfu32d7fxfnwp4oiRMiHQXVUlFijWF6epp6vU6WZQz7QZE1NzPFytIiQmmUlCRxEqwIhKCwOePhGDve5+HOGv/mm/+Or115k126FLbEmQCkehECBp2VzHSW+M3XvsTp9gpJFOGsObAD+5vARv0oAs4PAqo+DqZOSFfw/ZutH5rTVZWQjcasPXhIXhSsLq+AcfQ3d3j04CFnLzwRCD55QaPRDH9TBTVWnhWIJCHSmvqF45y4cJyiO8QXlqiesriyQpzEZNJTyspI6wireuLrHl73IcfdIhhkY8ZlQU1BDvT29+h1e9gHayAknUaLmYU5GtMd0pk2zaV5dKeBr8W4SGLi4Msa42mKDjPzDpzHa4mrAFx5oNr7+ZiMo7PHe09uCjIXpPLKeMr9Pmk+Rgy7TLc6dMc9hIowKNLmFNoqyvEeNgsgYcMbpLE4BAagKBCuxNsCG4+Yq9cpRkNmyhLvCprOcO/hGkMPC7OnsFt7SFHjuWeewjViWmPH+luXicuM5unztBdmGFIy2Nhg1Gmhpjo05+awNcX+cJ/SGzqtFkJKCutQOmLPFKwXBa7RYPETLxMfW0bUUz79v/lf8sH//Z9Qu3qLWV8y9p5MgRKKQZaxPRywne2xeX2deGvE+c++Rvv0ElGtgZWKsQsAZSIkg8Kwv7fP1u4OZ6MmNQfle+/Q/EQLIRtIr1EyprSGNFZYI7EmKEqlPMSYAslS45ylKDNkLoLKNk3xCEbjDGsMUDXvgpQXqQLZwNkAnGItvvIJdoSSS+Kq6yYhWDyKMKXDFGUgFCnBex9cpt/t8dorrwZsB3Hk+vHDGx8LTB0NhoeMRiogFb7nYiiFRGl94EEzMZQ9uuE33iHLsHmYmpqi02mTZ0NQsnpOibElUk9hZQLe0N0b8s7tDda7fTA2eIcUOc4qpCkYbd4nqcUsLh4n1xGbu9vceON1rkWaJ55/ic988TdZf7hBVBd05trcfON1+hu76OUT+ERQRxM1O+hWB1GMaDabyLlZ9nzBQhnAL6M9qQ5MWS8lRS542OsxGJdIJD5WaBWxML/M0uIyA1MyKnLGpUWRIazBlQWIwLKtNafxaRPtR5QNT+vMNDX9MuWwj93vsnb9Kl//1pvU04Qnz58N7EHrA/khUAk/1P2dAKqT8TgoepRpOmHiHV0Ej953yOasfv7Yc0gZGInGmu9hvobfPeyA/ziHsxYrJYkMhvCBXRp8UlzFbFRV0JLWGqmCB6JzLnQppSLSgfmZlYayKKHqpCulEQTmNdWtsw4VxUghuXjuJMLGlMbjMHhjwBQ4oUJquSnDvPUOZ0FKDQiiJGE86jPo7bCytIBUgpg4PD+CVquNUGCNZb+3DyKi3mzQaLZpjPqYPLBxnfdB1uRBRxqlWggcx4+fYGd7i9F4dNAAIRgOHFzkOAAtQicf6fClxTofEt9VRLvZoj/oMRh0w8nvJUIo0jRBSnBSBraIlNUmxYfP3R5aTAQrAQNCkuclo1GOlJpeX5HEgljJEA4GFTAB+GBNcHBLAGuTuMn09HSVDuhCqqpSeC/I8wJjApgamgeWoiwCoG1CWq2sSZxxlEXx1wL6/7DHpBmgtaTMcvLxCJPljIdDppotdre3ebi2RrPZpChzOp02Ski2N7eYm53luWeeDYyMrW1u3LvDS6+8zGJ7msvffY+BK3jmqad548+/ho4jVs6f4cH9+0RS8/RTT7MzHICHqVaH5NJF+nnG9vomiwuLzHc6jLMxW/t71GdmOHvxKYhiSgTKWdZv3ebf/s6/YuvRJonWzCwvsra9yejemH6WUTeKVpxQ78TMtto89fyzPP/yi/Q2thDjnDff+A4rJ08yt7CEl5pmYdja28ZkOe2pKbRWGO+CX7JWuNJipaOmFBGC3cEgBBwZGxZnEXxPi0ouEicJMh9z7MQKSkNSr5PWYo6fWGVrvYsdjvnmn32VtNng0ksvVnV12Jp78f03Fj8ff9kI57qn8mSUrrpKFczN1Hjh2fP8s9/5Y06cPEctThE+otcfIqQgcyPSJEW3GoAn644YlCOy8Ygk1mjbZnZ2mtTDVEdxfG4B5ws67TrYkumpFnEUMT83T/OLL7Kxvc3DRw+5eu0q94cZrdYie7v7dKaD7cnde/fxPHPQNP64+qHgkBpCOUsJynvMcESx12d5eZk/+sqf0d15wOkzy0TdfWbXd/jMXsro69s899lX+O1bf8bSwgVuv36fpFPjyr/Y4j/9e8c4s2pIfJ2xb7JPmzeu77L5wTd55YXPoRN11IDiCPt/8i/c//Pxgw0fMsOZHHyBJBv3uHH1O5jtErEyzxtfuYzev8ell5b41vvb2GKOJ155lnrU4MyJOe5ujbhy9RovvvAyWZHxB1/+fT73i5/md3/3v+M//8/+Ef/Nf/Xf8vxzL1Abd/nixQXOPXOWu9sl3/32Pabm6pyadyzMt3n+pc/w/je/RWc25Vvf/Dqf/uzfQU/PkjvFzOwy777157z6hX/A2UvPc/bCU+zs30fpFlKmQUJoLdngAd95/cucPv0knYUVCgF1FSPKMjSKhA/qGm+CT5oLoIKUgpXlZb75zW/QeLTG8ZXTGB8+IWsdRZFTekuUajpTU3Qf7AaigyyQYkwSdSq6lUJLicSxt7tNs91iamomBBf5UEfFCdTSOsY7oijl7OljTDU6FOMeBkMUL/PUc79E3FhklOdk+SgArz7BmBKlPMY4nBNI+Vcf37/Jw3tPURTs7e2zv98F4Pjx49RqdT7/uc8f7NV0FV4yYZc6Qt0pxaHCyHuLcY4ojqnV63zmb/0tkij6kFe9Mw6lNVIrrJ38nqDV6lRXGX/A9IvjCOGCvH2c5Zw8fRpjSsq8oLvfpdvt0mw2iZKYtBETNWoor6nLJs12nYUmlF4wijxu7JClRMUJuiJUWGOo6YjRYMjG9ibb3T2ysuTBvfukccyJEycYjYJHOs6TZzm9Xo9Go0GtltJuhebraDRiVBQHirzgR18FLAuPFopIaVZWV/nVL3yJ9dE+X3vv62EvoB2IsHf2VjKrp/mtV7/I06eeBh3jpEM5j0MdKAd+kuuGo6Dp46Scxx/z+NePS/sf91H9fl6q3vsDuy4ArwJR4s7NmyzNL3D3wX28kpy9+AT33DUGO13u3brDpU+9yHA4JMszarVawCYiRYFHWIs1FiscTgnEVIpQmub0NCZOkYD0jolX6gTAnPimfqQGRUBrZopLn36ZVrtJEsdEThB5ic1KbG+M2x8yfLRF98EjdByj4oioWac+O0U620a16sStBnWdkNZruEhhIhm8dWW14ngO/tlDHsrP9vC+YqcG0G2YjRnaAhtJ7HiI749I9vq4+3cYK01cT9DtNlFriv1ehmjWoa3JBxGun1GMhhRCEkuJMlBzDmkMUVGirMMPB4gi4Wx9CrUyzd5gRLOfUdY0KsvZvnqL0cCyurrCdq/LOBWcO38ONzfH+s3b2L0hp199BjnTId/eZmdzk/NpSqQUXkmyQR+cZWpqGuEF1nh2heWqyWnPHefiZ38B26ojc8fiuUv0/sGv88f/1/8b9V6wkhLSUxSOb9sh+zXFdJkyRPDBlQ/49q3rnLx4nk/Mr+KaM+z2+3jriOOEyJQ0M5hvddgb9vF7+0TfeZf6/BL52YsUuoZzFqnB2hwta5SuYnALgbUTRbA+uFYYa8krgltaq5HW6gFQHY0qixN/oHR3OnjOu1LgSovDopxACQKO5gkkCwyRrq731kFVi9nKKtF6w737D1hdWOLc6bMEwsOExv3DO2E+Fpja7/cCA7XyNvDeBS9GeeTCRwCuApAaEcURcRyHiS3lge+OEwJKD1rRqteo11JG4xEmL8A5Gs0GIBiNcoTWJIlgXMK1B5u0puewW+sMizGxtYy0JirG2O4e+5mhv76HUxBFnmY7Yrjb4/a3vonIDfHsHHdu3cJfNiTlGJOP6F5+ByUN+4lC1zqkrRniWgTZCNXvIxJB5oM3aATU8RRCUVjLoLSsbe/gPGgdIeOEdmeKer1GVozp9vts9rroOKHZaFWpkoZISYhidKONi+u0ZI2o1mB6aZVIeDyWctQlijV1N+Kbf/F1Uqk4ffpEAK0mi5cIHCg/kY9X/qVHx1E5v3MOZw/9oY5KLuBw4YoqBmaQtAcmAV4csE11JXmXSjIaZxSVx8sk6GdiIgzywIP0xzWkkmilcM6itapkShKpQpiUPQKmueBOj1LqIBU+UoooinHWgjFIqUJwVWmBAqV0OIGtOVjUtQo+JloozOT5sThrwFmcEKjKK3USfpQkCWUZPKvanTafeOVFpqfbaB1Sx0MXXqMjBcLinWdnawfnDO2pwDiVzhBFMXt7e8Ens0pZFfKwU2pMiZKwuLTE5T/9gDgJFgha6fBeAjJ/IIWYeJdIGTyLhXPYCgyNooh2q40HRqMhSu6GgoQWUni0jJj4xU1E84GlIA/moVICa0Ulry/IstBcMC4nTRS1tE4tTYkicXBtcTYEHLiJ1YStUpKFoN2eIkk2MaYI8nXniHREHAVGbmkKnHUHAWNCiANmRpbllGVJURQMBz+ZzNQP8cQ+dG6LqqlC1cgSKCUxRcl4OCYfjfHGsL2+zlS7Q6w1c3OzrB47RpHnZPkYYwzD/oBjK6skScLv/u7vcGL1OGefvIDtDXnzK1/n0ovPcebCEzy4dpOHN2/z8qc/xZ1rN9m695DzT5xn7e4Dbt++w+qJEzgXPHAMYSPVarWxgCkMuXGsnDjFzMoKTkYYQEvYebTBf/Nf/b+4ffkGBRYlJJu37jPxy51rtEiaDTpxjViF+fPLf/fXiFt1hvt7fPrznyOTjt3dXTYerhHV6nQ6Hebn5hgO+pj+MDRXmnVULSUvLamOQyCJs2RZhjeGREUoJRn5Eis9sZfIwlLrtOnt73L/7h2OrS5z+vRJdJJgnWXUHRDFChlHPLhxi51HG1TU7BBExWGSqHzsWP58fL8hDm7CCjdhqoYutBKeT3/yOf7pb/9Lbt54j6eeep5+1mM86JJlOXGkOXbuLNMtRbtdZ+rJec6dPUun3UJK0DKsY0pCmkSV9UuQQzlboGWQGnkPSM/c9DxPXpjnky89gbEJVy7fIblxGy8E9WMz4AyTlGsv/GNMkSPF2mOTwAsoFWjniDxYHSP2Mla/+YDFJKZ3epY/+tN/wwvZC6yeeZq+WGD21gMW18f4xTG/8MwrfHD/Fo3jM6xvrTG7dJE//tqbXPiPz9BWnlgXZMlDrr57heOihraTT/OQ8/vvP/6qmf2zsLPzj92EGs57wYPbV7nywbeZcoYt4cBbdnsZb771EEPMS5+5QL/oMdfo84nnn6TQi9y4scbN29d4/euv8xtf+jW+85U3+a1ff42vfO0PEbpDqxnRNg84u9hitpEzPdfi6VPHGGQw6t2n01xgenqG3sO7TLVmOLlSkDamiZxGeklt5SKInPUHV6idWOBRbxtRGE6efRlHjHRDht09vvpH/yMvvfIa7dlLvPnWn3PmwgskzVm08CDCRkV4g3KmauSH+lbgkEqjhKYsSrRWZLkL6hMhq6DJAmMt23tdhkUBWmCFJK51GBeavfGYWhOmWguMx5LRoMf0QhOpHR6LUBIpQr2hpEVKTaQi2q0GtSSlKEYgFM+9+Fma02fICkOWj0PicJzinURrGxRj5rD5/zd1/GXA2wREHY1G7O3tsbu3z81bt3j3nXf5+3//73PhiSdI05R+vx/qQl0HITClPbRKcoFYUVqDdyHzQimNFIJWq0Wz0Qi9LjgAuKIoCmeB/bA3qpTyIOtBSUEcxdTSAFY5b2l1AplmIv2v1+qMhsPwdVqjHOX0u7vU4zSE8KSKGBv8/YSjlmi8UngZ1mDnPWWW093dY6+7z7gsMN7T6/UZjUfYsgz7FmOIo7iqZ0MQVmjOSwRhv1OvNyp2VSBX+AmjWYTaKzRPw5pxcvoYf+/Tv0p3d5+37r7NyI7wyhDJlDSu87mnP8nffuY1alEDqxWaiixwRNnwkwymwofB08eZqY8/7qNugQM26gRAnVgJHoKq1QPFpNEn8Q6ysuDhw/tcvXqd/WGXWEj6e/u8/PJLdNpttra3OH/iFPWnn2L70SbjIifLC06dOcPO3m4gfSiJlkFVZ/D4SGO9RRvwxjI126IVpeEcqshyyssDJrsUomqWH65y/uDW4wqD6Y0o9wcYoYjaiqTTodFukdYbREKhkRhTYkcZZW9Iud/H9sfkez121zZxNgA+sdCkaQ053WDlE8+g04TCh/rB/SwssT/gcAJUYIAhgFE2JivDHnE8GuOyMWqUoQYjpHGooYTBEKaz8Dk3W2z1g49+nLbZWbvPwJUkBqZlgqZEiRxlSyI0WkaMs5JyuIWY7jDXTlg9e5J1nzEcDUmcJfYlDMfcvXGZcmGZY/OL5EJQKkXaaaHzknfffYsXX3qZxcYUxW4XncQkzTqF94xHQ+IkpV7TeA2t44uc+syn+NVf/btMHzvNbqRoGIXJPSd/4VO8srfGu//Pf0JiBGMk+4VhVFPkIibKY5wx7GcDMu25+kbG3PKIVWJqzRajosfu2NKMYuJRgXEjCgyPdjc4myb033wLsTjPeHoJoWS173c4EwhU3juss6HR5CylCcS2KImwWYYxIbQqzwVpWidNEzyeLBsfsFC1VljviOMYY3JclXXjSnuQ0+JlCDJUBEwBIYJ1o4fSeUxpD4Lvi9Jw+eo1lhaXaE8CBKGyW5ywH/79TqaPhXaVRU4URTgkvvL08PbDF07vCBemyq/IuhhrDZPezUGXSYVwBFOWRJTMdRo82tkPZ4CA+cUlYplUVgAZmU253c24vz+kNxyhyjHKFzipiF0A0KhF2LxLsdbH5iXOBplSWZY4O+TaN/6U1uwc47JECElrcYG98YByPCaWAkYeM+yhvQTTZKt/n8F730JIQ/vJSzS8gLGhHyuM9Ays4Nb2NoPRCKFiRBST6AaR0PS721B5p9ZNxqi7zWDYJp2aAwmJEqT1KXytTW26RktH1HzE5oN7zM+1mW612RcxG90xv/rKGba31vjqN15nYX6GTlODdxQ+dJCVDwi8RyOE+Z4F66jsP9zvDhh/37P4uSBJrac1mu02smL5WX8o2weIo7D5tM5RlqFIHY9G5JVhtrOh+2CtO/Cx/HGNKNYoJcJnlJuDRXlSBHlrEUoddMeLoiARwReVCmxlUjRW3ppZUWCNIStN6OBLwtzVCpkklCYndh4hEqzPED5HeIN1ButAawvECBS2NNSSBCk01hfgDXmeMzc3E04gUTFMhQKh8c4gRMnOzj7ZoGB+YZo4joMUSClUFKGGY/IsD4b/3jIa5nS7+xTZmK2tbdJaysnTJxjlQ27fXaNeq1FL66Rp/cDzVUmNlFHwxBKi8jx2UHlnWQPeBeB4qhXR8z3y8ZBeVwZZnnIYWUMw8R2r0qIdCBfC0wKzvQrOcAZrSmxpKJXBGsl45OnpEXFs0JEPpt+AqwBU52woLrxHSIFWEqkSqEKuvA9sAeE9sY6QwpM4QTYuqlA7DpgT1lq8A2Ms/eGIYfaTDKaKAzBpcmdQBlRhalKhZIlwhnG/S3dnl3qcYMuSrfUNWvUGg9EIpSPW19f54P336TRbnDl5CpsZygg6rQ6/9qVfIysKPJ4XP/NJ0nqNd999h6eeeorZ8ydptZpYYzj17CXu3L7NzvYOp86cZWd7l2F/QJQGuYeUEm+hmQRp3u54THtugebCPF5GxEJivGVzfZP/x3/5f+HuO1eIvArnXJFTSyNatTr1Zg2dRpRFgSAj9ooXXn2VUjjyomDx7GkkgpYQzC4uM9jvsr+7y/7OBrqekjRqNGtN9rd3yfdyknaLOEkpjYOipMxGqEiR1hJSpbGlJRIeSaWM6NTJypwbl9+nkdYRxpObHOsExlriuEaj7nj9G9/BDTKKwYisKKildUohQsCEg0yLnyOpP+gQAGFjq6rWjBAqqCJ8jHeOF557gpefP88ffvkrPHqQkNRTfuFTTzI3O0+73ub0yRM02zFpPaKoCirhPZEKTT4lJFpNvN5BCIvznijSCOFDQi8TFYgDC81YIaXjb71yhidPNWm1axhbMs4K7t5Z48SJVTwW8B9RlonHbgPYjpOUyqG9Q1no7u0jTUnkHGc6s6ztzjN77AzXZlbZ7EzhB1N88Y8us39/DXn2GEuzy5ilJbKtUB/lgwY9M0Ozvk3qPfX9NYbrd5h9/lNhE0lVF1h36PF6YK3y/eSVcFh0HqXfftQE/9nY5QV2pkegJuo0PALhSt7+xr8jkiM2R46733iDRuJYPL3AeHOf5dY8m+99l8Z8wuc+9xy7I0kkNOqY4otf/DV+9dc/xfUrD/Eqp9/tcu3qA77whV/jWGud00nM9MXn6HqDzi2RKJmqOdLaHC5TXH3zzyjLiAff/hZPvfQSOgp2NiiBoGB55QmmOotsbl1HxCULJ55HRjWEG9LtPuLbr/8pn/6F/wDrNV/76r9icWmJ97/5NV589VUazRaew2BDG94+zpVB0WKhM7XM/MIx3nnzTU6fOYVUEmcVTob6sZ7OIITGCVepbhxpLUGLmDw3RCnMTDdp1Vtc+eADzl04jdLhscHbPsjPvQAvQUqPEBalEnIrqHdOcvr8s3TmTpFXdYP0ijRNK2VMVQM7iRdVPeD1gRjmp2E45xgMBnS7XXZ3d7l9+zbXrt8kNyUz83Ns7+1ybDRipjNFp9U++L2iKDFVQKiOIwbdAcYYBsMhWimmpwM7SgkVaF9SUpYFrqqzhQzeqAcqRg73GxM1nRTBiihIpgPbT6so2IlN/OuFwDlLnASQU8dBkdVyjp2dHe5fv8/MzAyzc3OVagy8KxFKoqVknGVsbGywu7vLaDQKwb9pSqo0p44dY3FurqovLTKKEAQQWAiBrcJYi9IA4fsA/LkDFeYkHFUIgXGV9Z1z4B1KOC4eP8v//Ff+Pur34f1713EK6qR86uLL/NZnfoVa0iABRAleSowIDCvhJfCTGUD1ONv048j54cOknqO/c1QGfFTmL6UMNoHBoRfw7OztcP3qfW5dv8ftO9fZ2dvg5RdfIU7h/qMH4DzGWs6dPM3169e5fecOF568SJEq1GDIxs42rekpTpw8xXAwDHPCS2SlG5Y+BF4jAojfabUrzoIPvopVE1JWRAbgwA//6FIXYjMBZ9m+cZf9q3cZtVrEtZStNKU9M0Wz06bWadKan6PRbpO22rSW5pFaIr2gHI3JewNsP8MPc7IsR45yfKyRtSTYDRz5s9If8Gd+PgBTNVLwnsjC3tY2GEczihhmGabMkFmGUIrMGxJb0uiXROMRYm6KQhqmnMMqTS4FeS1hqr3I5p1NxkLSEp65RBN7Eywek5R6rHHZiLy3TTnYJF1Z5OzZE2zv9mmdXSavabr7W7RGBbVhjnO7bO5u0XnlSXyrTmYFraRNOfZ079zmxIXT2Jk6cipBi4issOwPBqS1Ok7BsUtn+c//T/9HarqBsQ5lLUXkqUsoJDz/yVfZ/+d/wNbOA7pK0/OambEnV5b9uKAcGZrErEaKjUHJva09ZBKjhKV0Oc0ywVjPw8SALmnkgWiyPthm+d4a2XvvEr+WYr3CS4VxGu00SlmMLQlngcERmpVSKSIR7DDLoiB3Bc4bEJ4kTkmTGCUFWZaRZRlaRkhr8daQiCSov4XHSEtpLWKy/lhAJTgrUCp4tntpcKrCcUqPcAorPPfX///s/Wewpdt534n91lpv3Pnk1DncDjcHABcgAZJgEilrZJbKmhnNTNlSSaVxKJenpqbK88F2lb96xp9sf3GVq8ZTM1OWRI1CSQwaggSIDNyc+3buPt198tnxTSv4w3r3PqcbFyQIUOCliIVqdN99Tp/e+93vXut5/s8/bPHB9eu8+vJLWG2mbKRjHfZP1qj9SGCqqDc7i28ypgE/x70yecKvcyr5lsITrpVSntVaaVS9CRnnWJxbIBKb5M4hwoS00UEXGd20SZVVjJB859oH5FVGKCxGOIQNsSoilZGXcQqLtppyOMRUQ6SucEIh4yZhq82J9XUOd/Y53N5FKh/2tLy0TnttmagZMOnv8ejuXcb9e7AvqFSEEI5YBmz2R6w0U6SGvlNIp9na32frcEjl/CRVCEkUBgghyLMMdEAcxTSjiOFeRpFllGVGa24B2+wi0hadXo9mBMPDAbe3HnJyY52lxUWyqmQ4POTMxXPY1hKt+Q3efvd1sjKn63oY6b25pHZooTxTRxcYHj+snrRgmHml/hA5hnXOS2Pq1PVeb44oTnBC1CxOMZPJO+cDyADyPCfLMooip6qNi6dsxp82mCqVB/OnzNrp850m1Vvrmag+YMkfvq4ofJFtPcBfTf0zLTOZUKU9OzebjInjqJaVOxpAEE39/Pxr1doiao9hZ494aWVZ0Go1PUO4qgikpChLH2ZV5LSaTe/Ram1NmzcIDIPRIYf9Q5bmlojjCOr30xpNkWdMJhlaV6xtrNeNhvO+pMLx4Qfvs7K2RpzGrCyv8Mdf/ypxHNNoNEjTBmkSkyYN4tiDq2naIAhDhApABj4MzqMQGOs3xSRN6SnJoN9nPB4hBUdbkXNHkjHp2QZW+GZGSOmBVKmxpsLoirIsENIzu50zGFNRlbm//u6oacK5Yww1hxCOSgriEBYXV3iweRupvMm4lIIoiogIcFicG2OMQ0rlrRZqQHUy8dYlOIuSn85K5HFXl6MKzrN8JQjhw9KMYzwcMjrsMxkM2Ti3wmBwyHNPP8MHH35AWZZ0u11OnznDr/7qr1JMMt5/+12cgCtXr/DeO+9SVRXrp056MElK7N4e585dYDAYsbu7x8L8AlllGI2GtJptPnr/Az7+6AZJknD9xg1efPEFdva2uXL1ClIFVEYjdEjS7dLszRGq0DOahN/zfu+f/nP2HzzygzlhSBsxc70OcRT5MIaiINEBLRURBiHjfELY7dBbmmfwYJv5tQDbTdHCERvPSF46sc6jGzd4dOceqyc3YNGSLnYRxjLaH+DigrjbRskQEbQwwkBRUhqNBeIoRhhH3IzJTcWNazcwlaWz1OPg8JBLTz+Ndo7haIyrNAfjXSqtiYKAqqrIdUmTBsq52pPYF9z6ieL3Z+vPvo6HIXrblZC/93f/E+KkwcVLV1k/cQIpQAlJEqSeBW81eWFweKsXJaVvh2r2kx9IeHBcyNpGxPhUUal8rSPFUbietwkxM8uY/mGfMPbyop2dbTqtlPmF1hPP/NjkW4gpIonEEtmKTMZYJIEJUKHjKx99i52nenQ7Te7bQ7Ig5GGvyYMg5E67xbc+e4rn37hDez/nwuppdqob9FZW2X3tY5yrKMQKDwdt1ubbuNzQEhY3HhJ0Ogx1Rccm3sriWPP6Z1/Hp/hP8nL+ii0nAA+6uLqcFjbg9s032dn/AFEGbD3aJwgSDJqd4YDuXIOw7VjurtKd63Lr/RuEvSYLi33SwJFSMN8Y8bf+F1/i7/39/4B/8U9+m1df+SxN2aep9mivP81Ah2jZxImSPB/TClJs6WhHBUl7QlUmnDrzLJUImQzuEDW7UHnpvImbyHaTE8nzOF1gtQKl+eDdP6AoKz7787+BcQlJWHLizBqb97dY3Vjlq7//L/jSL/86SbNbA5iWsj6Zfbist6pwStFenie8aXn/3bc5e/YZgppJcmADghTisEnqUlp2TNiUEIYcFANePL/O1RPrJEHKd7//DmfPn0YEAikSlCjAGQKRAtbbG1mDCKFwhqjdo9c7x1MXP4O1DR9yWY4x1hBH0awGnrLepJRooxkOR6Rp8hd5F/25relnejAYsLu7y3g8Zmtri4WFRb588jStVotmq8VwOKTf79NMU5I4wTqHrrxtVxhNh+IeRByNRnzzm9/k/Pnz9Hq9I6VhHVgij/WHlfGMVoTnD/n97qhV9X6qltFkzP7+PtZaTp44QSttzMBDrTUPHz7EOec93tOUyWRCFPpaYGlpidFoxM2bN+n3+ywvL3trNimpqsoH9joPio2iEcPhgGbSIkkS1tfXGQwGhFHo/WFrpZoSkjiOPTtV+ddjtKasa/JpHwE81td4EHDKUHReQeS87P+ZM1f43/ztv8cb77/N4bDP5ctXObG4RidooAh8uIoU2BqCk84RTBVZdT7Jp339afL+T2KtPnnuTHuz46DqlOWOi3jvg4/4b/6b/5Z7d7bZ3ztkfiHl7/+D/4Q8zygr77dotOHu3btcunSJR48eoY3hzu07XLh4gW25gzaaax9/RFnkXLlyhW63Q5bl3v6p7t0QliSJabXaPxGYLYQgryq0hN7KEnEYIqxDZiXF/R24s0uOYBCFRGmDqJkS9ZqkC11kKyWZ79JcnCNYXQYpKRUEzud9uHaCwYL0VitBTU21Ygru/my5eoTjJAhtmIyGOOkwylKKCgKfpSKNQyEYKx88mzhBvDemcegV0pOFmH6eY40g6C2wY+7SWFpiMqnYHI1ZaDRIpJewK+1oECK1ASUZbO3RH0/YWD/Fxtoaj1zFPXtAYyFiNCeZHAwZDnKiw5xwa4zrNHn6hSuIw4x9mTFQGeHmFh2jMIvzBHEDV1SUkzHNsIUSEhkElLbypAMHwklEI0E4zXjzIS3jsI0WmXGYyqKVxcUKW+Q0woiGcZwTESfCkoeDO+j5RazJKKzFhBGRiJCTgm6niZuAzEFozX5+H9mZ58K5F9hLA8a9BBUYVFlgp/iE9UMxqRTG6JrAZAlVSBBGFFVRZ5X4sziKYqIomn3+y1rlcJwQ5/+sAV1bCHCkYK1DpaQEJQPCUIDTaF36z3Gd4fThBx/wzKVLNNO4vlvE7P9/0v32RwygElTap+taMZ3SeWrtbONH1AwPQVmWMJ0gyyMfFGctUgQIKu99ZCWNpMFco8ndnV3ai6uEjS5RGqOEJTMhH+zucX//wPsp2Iq02aLVmcOpmMNiRBTGuHGGHg+QTmOVQUYJUWeBdGWD5vwCg71t6HS5vH6a+eUNFlbWWT+xTBBInAqoXMDN2x/x9ld/h06pUVGHAgPFhK3+hP6kJFGKihIxmTDMJuRWIKMEFYQoJQgDzxqdXhthLKGSNOOY/f0dhC0YWM1iZ4FWbw4VSvYfPGDv7k3OPfcMq2fPk7mKbNLnwsYJXHiOXI9pnDhPfOsmuirRAgwhzmkPvCFQzuHtruUMKP0kLxshhPd4eeKAm/qgTunQpk6Aj+KYVtuzuYSUGOtvXClk/buXVVdlRVEWGK3RRlNVmqmP6097Yy/KEiW9b6+aHs71/Tj17ay09r5H1lJVFbH04UVSSp+YOJmgtabbauOM9cFVQJI2mEzGMylTnudYBHHT4lVifnrpaeoWY129AXhpU6PVQErQuqzD2kLfqFtHmjQA4dndImAyyWg2e0wyP6Xp9XqoKMTVr6UqxzTShChK6PZ67O7t1p6pgij24VqBEKyurnLv3n0Wl5fozS2wt98nLycEYUgURzTjhEbaoNVs0Wy16bS7pI0mcdokSduEYQOhQg/YS5/uqo0jTht0hWQ46DMej2fAurVmVvA6FUDto2W1QUqHkRotJLqq0FWFNRqrAzSgnKoHEwDKN6nC1aGZNZfYGRA+qQ8lIAy5fOkqk3Gfg8MdTFVR6dKzaa1BBf498d6+UwaKHxoEaTrbnz6tjdQ0BV4gZl6cOD+Bk0KgFJgqpxyNcFXFuN8nUgG2qmglDYw1nD19huFwyPrGBh+8/z6D/UOefeF5Lj/7NOUkw1aGz3z+VcZ5RlB/tqeflzRN2dnZ4dy5c77xMZZs7FNQX37pFfKqxDm4+vTTSAEnN9YI45BMV1gVomVMa36BIEmRqFp+ofnt//5/4I0//AbFYMT8XIe0nZAKhTQWhSOJYhqtNtlwhMERN2Je/sLLvPqln6MVx7z2/nvc//gaT7/6Cu2lebSQyCSiEpaF0ydZmF/i9rVrbN6+y/zGKgsry8SNGAlM+gc0u11sIFCEmFAwzCY0Qu/2rwPB7uEuD2/dJQoSVlfWefRgi8XlZUaDMTIMKSYFD+9tsnn3Hku9BTajCOusby6dZy9YWd/CeC8r+VcUa/rx1w/KbmbSPyEQSBppxGdeeQEZJBhdkDQbCAtKgZD4vUCFKBEgkbWH9pF80IfseYzT4vdP5zxTSNRhbl5hcVTMWWcRTtBsNhmN/Vngifw+jfpYZ33s+R978+vHLVBIWSfweo9NIyxvfPQOX/ilL/L6a69x/2CbPIId4IAIIyL6jYhxr8PiwYjzrS7fFobJcIwwllRIci2Iy3lMFTHQJY1ogzBJefHVlxlPCnqdzmMD1yMt5ePP9E8/u4/Lo/iEP/9Vaes8iCqcA+F9vTABb3znjxju77F5cxcpJKurAe20wfL8CeLIoFTBg/0Dbh/s0Ww0WGo6st2MVrrAW2++wYmNFW69+W2eee4y//Pf+Dm++s2b7D/aJpFr2CiiFYIkoKrw3mISwobCOYkrA1JRMN67zeKpZwjDiFjjbYeSAOkEcVViRYCgjTbb/E+//y84d+kKK53zSNOk2QmRYZunnv45kN/l4w/f4ezVZ/net77G01eu0FvZIIgjhK2wVvjPkYHSOrTLWFlbZO3EaV7/3ndY6PVotVaRokEUDmi1ArrNkHYjoXBtcldhrebl587zxc+eZefeNlkhOXvpPIECIQKUFEiXIIUlkA7nYpQAFySooM3SyjmWVi+QJD0qF6BNRVHk6MoQp7H3dptZUB0BfEophsMBy8uLf5E30Z/rKooCIQQrKysEQchTT10CRG0D5YfJrWaTvb09Dvt9lhePhvRCgKn0jIQRBAHtdpsvfelLJEly5H9X11JKKT+YwZMK/Dr6/D/ZoHp5vCPLMj766CMebT3i1c+9yuULFwnDkLIs2d3dpdFosLS0NGuujTFUZYlKEpRSrK6uUhQF1lreffddOp0Op06dIk1TtNF1UKyg2WygdQcpJWGgiJOUbrc7+7lFzYSaZghY64GqKQHleJ807SGelKOLWqZ/fM+z1tdt661VFp7v+fpAKAIEgfUhuUjPtrJ4UEfUW6hzjtFoRJqm/5bvlD+/9Uly/uPM1Ce/b/r4cdBy+t9KKZQC6zSPHu7yzW++zjvv3GQyrlheWOA/+z/8p2g35O13vs/zz73AbpJijGFra4u1tTVeeOEFPvzoQ4IgYHtru05DV+R5xu7uDn/8xzs89dQllpaWUEpRlHomQ46T2AfqTlmoP+YauYpodZ7VC+eI202SZoNYKEILYlKhRxOqooRBjhtmZA93GT7YxiqBEpI0SRFJjOo2See7yHZK+8w6QTP2oI0ArY6BqD+rL2dLODGzRSuKnLLIcMKQV4ZxmeMihWqkMIygcAgrqEIDDuJSk+YFobYYSubSBs35JUxlIBIki03G+5pxYTEyIS7HNJ2mGbeoBEgLkXUkUYTJNVvXb0AcEazMc67TYE2FbFeWcZ7R6nWZCxOuvfcu7bUlljopYZ5z+uQa94ohenuXwVaf+PRJVs+dR0nF5HBAO00gDLHCoRS1SlkgnGTiDEIbytubpFWBjiKSynEoc6I4pDIT5lBEVuCkZcFUnGsr5mVE5UqcDMhVRFE5mmFIGKYM+mN6jSYai1WWzOaY27fYu3OH8vJTOA1OSWxYeYFtTdrTbkpIkIDE6AqJ91ANnKmtIS1FkWOtDwOUUs7C/8qypKqqmR/30UArQAiD1rXi1dhZcKIPHw9wTuACQRQ5iqICZwnDiCzP+ejaNZ5/9pmaCAbTWuAn7dF+RM/U4WxqJKWc1c3GGG/yWk8fpynoUilio2eJ2koe+WcqYZGyQjtwNc2+00gRYcDi6bPYKMXFMaMyY7sSfP/2ParKS0DjZpO42WFu/QRp3GR49xpCRUQhVHaECgJcu03aXaS9eILe4jqHhwPWT15i/vQZjBDMt9s8eHCPO2++iR7nuKJERYqVi5c5/9wXuPOdbxBVOardwglNWYwZW01lwJiKUgu0BWSAUAopJa1GgziKsEJ4kZ+QNZsOOp0ug/1t3GSACSIkoIsx2eE2t999h42NdToLCxwOBijlp6mtTpexdZRSIuZW6S6vE8cxDgtWIayPFYys3wAqpZDTm7h+P45LK+BI4nw8MGq6/CSgpCgAJJ1O6anZwktjVBjVHrly+sO8wbP1AILWnlKt658/C6oSP11mqlLeG9M655+rEMggQNbp7mEUMR6PZ7KdaUFVVVWd5lkC3jfWyz0hDML6OsSAI89ztPZy5qL0m8fRpawlQQIvPZKKSZYx3+vhnEXrujAzGqg/S3VhinWEUczewT6lliwurlJVFY1GgzhMMJXlsH+IVDHtdsfbDAhBFUUeGNZeXupwM/n3+sZJNh/toCuH1g5rBZVxFNWE0XjMgbMEUhFFMa1Wk7m5OZrNNq3OHN3eIt3uEo1WBxXECCHRoqIyBqkhbbZRKmA46JNlGfv7u5RlTjNtkMQJcZwAFik8mClrZpgV3m7Bmqr+pRDCei+iWrsnpTkqwoVn207txG0tpXIIqiomSZosLCwyGh8SCIGUirIsybMCqQTGmpqRSm03AGmj4cFc570T9VRa9qlbx+XCXo4kpEAqiRIWTEU+GSCNIxCSOAh5+OABb772GnPdHi+88MKsKdjd2ubM6dOkV6/w+vdeQ5cV5y9dJE4lr7/2Gr25HmcunCeUiizLaLVaHNapuu+/+x5nz5xhPMmZm5ujLHKmBt5OwOGgz4fvvMsLzz3Lg4cPaS3OkzvH4voKYZJ61UL9Wj56621uvPE2c0FI++QJJqbEOUMzCGg0Es9sCQK6Sws8/fILrJ8/w+LJdZJ2i3azhZkUnDhzEmc01999l6dfepGy4dk1LtdoB41ui6svv8TO1hYPd7a8xK+R0m00mezsYcqS1sICDkEoQxpxA2ENSkp29rb56Po1lrvzSBeiK8PlK1f55re/w/mLF0FJ7t+/z/LcArcmGXKUE9S2G4FS/hMoHjNm+Fmh+2Otx66gb7rs0Z4AYHRJVeYEKFQQIJgG4Pl9zeppaJ2ZgQNh6FlXXkGhZ0WaEP68UHVH6/CM1UApwDfzQgqsBomgrI48wpEKpaRn4FN3NzPA5jiD89iZi0TLgNgahLSU1mJzqAYl3/yDP6IRNYhdwSFjRhPLXipJgLaW3F3pkS33KBI4HO1zsDvChSFWWXpCsTDfRExiFganOaiWCBorlPmY/v4ua8sLs3NHOPfYVf6TbtM/mcT6gxYGf1WWFV7qF6DqE0owPthh8+O3aYiQTtrk7OkTdDoRDk1Wabb3BzhpWJg/Sy9OOLW2ysO7t1iYX+TR4YiFhS7NUCPGJd/+/e/QXU5ZXOoyHAr+8LtbPJc5nrtaIJOAMF4A16SqKiJXW1spxzgbYdAIM6YwCySNOd/wSEFgYqxrA4dsbb3J3Ts3ufDUS6TtHlES0U5ThPKSfW0l5y5dodtp8Ye/97u89NJneP/99zk9HrJy6iwyjFBOY70GH4PxQZil4PSZ85R5zr/8H/8Zz7/8eVZOnCYOIwhaHBSWvaKiP5TMtRNefKnL5166wLf++H1EqDh7vunDe2SACiSSAEmAQOMERHGMEA2anZOcufg0Mu5iREBhHMYUVLUveKPRmtXiT9bDumZRip9ynfrntZ4EppxzMwAwjmOeHGpM62OUZ9P3ul0frJRNSNKUyhhKUyEtnkUU+twLpdRjwN50KD3dS3Ud9GHr34XwBITjDEspBBb//MbjMe+//z4nT57khRdeoNvtzhipW1tbOOdYW1ubgZu2fi6TSUYYReC8JP/EiRMMh0OMMdy4cYPh0Af3NZIU5xxbW4/YfLDJ2XNnayDUE3p0VRHUDXocx7V60nv8G2d9kG39Go+DqVPwdBqsDLUk3fn61CJnNh9Tv38AJSOEMUjnyR0CX8NZZ7B+hocEgnrgNw1U/bTJ/H/U9cPYqE8+dhxMPe6f6v33Cx5tbTMcWoaDnPGwwhjHf/wf/0ecPHWSr/3x76IkpEnCwsICUgi2Hm3x8ccfc/HiRc6fO8/9zU3iOEYAURTRajYpW54wc+fOHR4+fEi720Hg1ZYnTp8gbdQkC+HJSerHfP12ktN/tEuQTGj2uswvLxK227hGSmN5HhVF3hPTCMpJRjkaQ1FhxwVmNKE4HFOOMqrtQ8yDfXQa0l5aQK54QCkyfnila96Jcl71/FfwCP6BNQXBpbNMypz+ZOit/Kz1ftAW0uVlAiUxD3eQxZhEBVinKQOYOO+HGg8LZOVQkcZUBadaMbFy7CApO/MEi4vc+eC7LHYaBGVBN1D0nKTlHNKAwRIEEpNllLfvITJNq9HhXLND2GywX2nK0Q6RLGg3I7LBkA+u3+Js4wqLYUAa9xhMcuz+AeXygLTbxSHpTwra7dBn5tQ5KQIQFqxSRJmj/9EN4qqkCkJ6c12ynkH0R7R1RTMIsYWmaEqkFDSEZSV3lDZkX4AloEAwLHOC1KGVIasy5lXIqBohhKF54Ljz3veZP79KWIboVuLD0bISVytapfQ1kZR1oGFtxaFCr15NksYMl6qqEmf9uTUFVKdDK3NscDdVRitV57MYrxqQcprFomtsx+/TYRhiraPQFVb6s/77b7zOmTOnaXdaKLya7U8pcH+k9SOBqaPJeCZtCMPwB1iNgC/OER7AUpK8yAnDcHYBgtq03FqBCiqMFVinAB/8EzWbtJdXmBjLvf0Rw/4+7233eTAcUxaaXtrBNJp01k+SNjtEgaLY2Sfu9DAWSGLSxhKq0SDqLjG/chJpJGfObTC3voEzOYcHu+xmY86dPUtw/jx3r3/M/uZd1haafPTa14mTBuvrK2ze+hBJjmqkREGDPHdMSosqvW+ECgOmcuMwVEShl3kGSeJZSaKGvGVAkAakjTbZYA9ZVZSHewxsxaM710FDtPQid3f7tKOAtfkFojBlazBgMBzQ6s4zLi2N5Q1Eo+sbflfWjFQQTlM5MARg9WMH1vFUxOPg6tF0nsd+z7OSoigoS02cpMzNL9Bqd30hIyRBGCGVxEmJFMpvWM6HkVln6hv7GFhrxU+9EAiCaaPs/VxFbVEgVYC2jqieWoCefYjLWuava1Zr3GgQRRGh8huAFnhD+9qQPwxDpPKTS1fnMWtj6iLSbyAScM6QZTndbs9fj/oxZwwIgUUjgwhdaVCylm1LHjx8xHBYcO7cxfo1eUZUVhTEcUi3261fg6sn6z4YqyhK7w2LxWmDSxRnLlxkbnGVTrvNaDzk6pVneOOt17E19d5aTVkZinLMcDRiZ3eXJE3pdHssLKywup6zvLxOp7tMFHlLDSkNlbEoI2i2eyRxwmjYJ8+GjAaH6DKnSBo00gaNNPVFZu2ZIpXfzLSpsMZQlYWX2LsAqYKamVrftzUCZY8xAMBha2PrKXvVe7AKb3wd+D3IM8oMVWUJQg9a51nu2SlCEIUKV0uDwNsCfHrX44CMZ5N5k+7R8BCrK/JRhbOOhfl51ldXeemFF6mqisPDQxCC9Y0NJuMRH334ESdPnOCLP//zCCUZjcc02226vW59ACn6gwGbm5tcunQJIQTNhg91mpvzA4GizMmyMaGSiCDEakOn2+bq1auMxxnWCiZ5xdz6OnGS1owLiZNQZRnvfvs7nOh0GRUaJRydNEWWGicEudMsnVrn4rNPc+nZp+kuztNotymt5nBvj1KXBM5x5uJ5rDWM9g748J13WTt7FrW4SBJE3o7DGIIgoHdqg7lTG2AM48NDBnsHOG259sbbbJw5zcmz58iHQz549x1KVzEeDxDa0J7rkaYxK0snmIwnHPb7vPzKy9y4fZvBaOjDMIqSOIy4+fH7friiJK6ocKnD4tmLrr5ngxmu9rNK98+2PpmdOlVgNNKEQEo6rQ5CBTiD92EU3tYjjhMc1odHSEWowlrBMVVt+MGbEqCNn5jrOrRQG00YhHWNY5C1D5dUAU7775ka7QcyJE0TtK5queLjr8E5fx8cnYe++ZbWYZRDC0ugIka72xSjAd96822ePX+W5YWcdmOO+4cTqq7hZJnTPhxgnj3PbjvmYZBR6ByzO6QyFZomy+cWafU0q/dOsXnzAncunuJ+VbLd3+ef/PZ/x//2H/ynvPzyyzOFyvE64c/+fvxpf+/f/SmCdJ7Z4nOZfFr9V7/yT3n2uRPcujGht7KKpEEpJZUdsT8ZMMoczkCRjzh3rk17bYNTz73McLSPvv+AzXs3kTonCmNOXrgAImMwfkS7XfE3/s4v8J1vvMfw6x9w/vwyC6sB7U6X0gaUOsdhyMcjqsEDHI4H93Ki8YTgwgJJ1CN0CiskVk54/9t/SJIKLj/9DP2BoBUkpHGBYYSwAoUlRGGsYHlhjV/+8q/y3W99i6euPsv777xJnmWcPHcBGh2E057dbR2uAohY6M3RX1pnfnWPb333q0RvhYRBl8MiY2//Ia3EcOZCm1efP02vG3H/xgNUpFg7uYRUJaFMEQ4kGilSb3EjA8K0gXYhZ06+wNqJ59FotNSUxYiysujSEKqIRiOF2rPtScDUWouutA/a/BTvy09+Nj/p82qMmYGo0++ZAqmPpaPXAx1vdwJJkhDHMcPhkKL0djfD8ZhmktBqtjB22qj6a2etRVsDAlQQoJ1Xd+3t7TE3P18Hffrex9Us1VAFJHGMUhJjNIOB5u6t25w/c5anLl3CWJ+kLoHDw0MajQadTscHg47HpGk6YyYZrZmMx8RpQlVVCCFotVqUZcnGxgbb29scHBww35uj1W7xjW98nVOnT+GcI0l8mFCeZz6puvZJzWsWr5ASWYdBFUUxI0xMe9wpgHqcQTmzURPTKzv9n0+Z9soUh1QCJyVe+eqLgXo0jnJelaCEROKv9aTMiKLoUwmm/klnxZOs1Md/2bpXdEwNPj/JP1UI75t4995dikqjZItr164zmWRcvvIUX/zS59javuXP/iBgf/+AkydPsvXwEU899RSbm5t88MEHXLlyhVMnT9IfDmiqJnO9ee/FqEJ6vfmZJYRU/p5NkpTe3JwHptwxib/7k1/zD70WB2N2372OCFKKOGHoJHEQEHdatBZ6pHNdWgs9wk6LcKlLc3XBByEjEFGAkQIKA4MJxW6fiS4JFzpUwuECn06hvJU75tgt8um6W/5ilqzfL+UEmdGUShIY4eX6VuFUTN5K6S6cQTcTynsPSPKc0Ai0q5goS+AMUSUJnCMpQ0QuacQBZVbSQKDSBNWIyaKQZH2dYT9juHtAgWKsBGmkiI2lrR1x3acmlULs9Ck3t7FKETYCMr3L6bk20lU8HByyVY447xSHb91ELS6w/tQZ+lHAndsfs3ruAt3uAv1xRtxMkVYR1vRKh1cNYy1yXDJ+sE0qIQoClk+d5Llf+jm+8f/9R6w8khSmYNSISIWmEUSUZUE8VlQo+pHmUDqvDKf28peSyJYUtqIbOCJrifYHBDfukN+4RnjhKlErwYgAJy26ZnUbAUZrBKIOjpa1H7XP+okiHxKO8GdDNQ05DwLiJCaK45la4bi6eGr/54VV3uDSVLUftzyyXjlSEIBQgqIqiYOQqtTcvHOHZ595ZvaB8WaMP1m9+iOBqePxeHaYTA+xqfTiSaajqFlIUqo6IVKhpJqBQiCQgUWiME5inWCSZ4RpTHtunvKgzzc/+JCJKdjNDXmhabbbOCvoLa4yv7CKyAtcJEBocp0RtbuEkSSNU4hbJMtrpEsrdBoxkRTsbt/j8OEjrj77HEmrxd7uLv3DXQ4GA9qnznD2mRdYubTDe+98RH9/j6A5jxsNKTWEaZMwlGjhZcPCFjUjRRDHMYHyTLhQBKjpG6hqUDny0+GkO8d41Cfr95EP77Nza0IoNdH8Ets7D5kXMen8InmWcbC1y35Rsb6+Tigdy3MdYnmOWwc5z8zHxLJAyACnBQaJEQZ5jI06XY/J/Y8d/EcH9LGpofVT2cFgyO7ePpMsJ4xizyUT0HSOFEcYeZNgUTdVvsjyhQAwmxp4v9DHvYV+GksGgfc9rSryLPP33Aws8xJOn2ZfS4e0xlQaIofVHoDptDuPDQi09T5S2USjrQUhEcqzmajvcWssSPxm5iqE9MVmGMR14eWBd44VF9QMTRmoOvU2xOGB2f3DA1/IOYtUEUJI2u0OcRoipcAzNH2AVqvdoqp9vzrdDlJJ0kaKU4pExaSNNhhHmqT8nX//79QT/IGnwQuF8CkoGFtRVpqiOuRw0Gd3f5/BZEKlKywhS4srBGGEtSGgPegeSZqtDo1GymgQMhkP/ITJ2doXtZwxGAMVIKWt2X4lWnupvw7qEJiZLYMCaRHCMmXvqnpznN6rQtTfB2xuPmBnZ8e/38ozqa11tFpNVKDIsglVnvsQinpz9eC3XzNbkk/tOiqPvN2C8AEL4wFWVzhTcXCwR1VqTp44yWA4pCgKRqMRrVaLubk5+v0+o8mEs2fP8mBzk9ffepNzF86zvLRMs9nk0dYWnbkenWaLg/19zpw5zeHBAUkcMx4NabcabG8/oiwLgkCRphGBEozzAikDCm0QQcDy/AKlg0Ipmu05b6dQM1KMFGhdcu/j67S1Y25xnsxphFQsLK1w+plLnLpyieWTGzS6HUQc4OnEjsH9HdpxjDaa3tKCvwmcJYoj2t0ud27eIZUB9Lrk0tFtNEE7qrycsWfDZpNGu40whkkx4fWvf5NqmLF86iTtdouD3W2c0Vx55ipp3ODmvXusrJ0mbTVJWw2yPOfMuTPcu3cPU2my8YgqzwiFIEhjkIIPX3uT5195hXC+g7Q1Q14IlHGYv5zkp7/gdaw1qOX4HpP2jzfTJlL4hN96tuf3ZetllkEY1KmiIJAURa24qPfW6Q/UxqsojPaghIz9eTFtqI2xGA3WWcLaj3s6PGs0GhjrlQhZliOQ9Sb1ZBP2BCiMxY/kAm9xYi0Hh+9zcinndb2PdIJnT8/z3Xdv0b77iOfZ5QR9TquA7UBTdef44O33MXcfsb9zn0aUEuuAV19+mvDeiNs3Vrh54RU+SBOkFqxEKZODPl/5w6/wyiuvPHYdfzzv1B8FSP13H0x1BB40sQbIefftb3Hz5hssxBWhjBiPD5GixIqSvBiQDSqqTFC5gKjRYu7sU5StDgMR0l06zcbKVU49PWBn8zrbm7eR0qJshDWrLC9U3L/9GidOtXA7i/z2f/tNPvOll9m4MKA11yUJIyDA6oo7dx7QHxxytTVH0hvw4MGbPHXh8zjXRVd7XH/n37C4ukR77gzXPnqbq1dfJFIBphohZQEqxNWWOoF2ICQLy6d5/iXNtQ/eZXntBPv7Qybjdzh54Txpp41QYQ0SOYyVmCJmde00Z4shQQr9nX1KPeHZs0tcWe2gC1CRRMSKNz56hJSKpdUGSSgISJBolIxQMgRVIcMYFYcYEXPl6V+g3TlJicY6ja4qiizDaEcjaSGEwhn/KYPafkXKWQ9SFEXNSg8ovCTrL93ypIFqVtM/DrwdrxvqP9eEl9m5VMsmG40Ge/v7HA4HvPHWm7zwzHPEpxKkUt7WSggfJiyPS/p9zyGUpN1powKfieFr/hpUtI4wiXwoLL5Jnpub49XPfc4DpMLbQwRSMh6P2d7e5vLly3VQmJdtTwHVNE2RSrKzt8vS0hIAZVnO+tBer0dVVezs7NAf9Cl1xdWnn6bZas7yE/x+HswAWq11zeD11giP5SrU4PS0ZzoOoLpj95KrVXDTMdP0qstpaz61e5FTBqpDCl/DT3MB/NcVGumH43k+Y+t+mtcnSfqf/DOANzLwKkb/9SkR5/GfN63Fb968RVEWNJodHmw+5MOPPkQqePXVFwhjw2B4gHPQbLR5tPXIM1OlpNPtsrK6wq1bt3m09YjV1VXm5uYo8pLDgyHtTotutzd7n5O04Rmo1jE3NzfrY2dno/ukce6fvoQQWAELJ9foxA2vLjSCuLLYSlPu9tEPD5gohQwDb83WSGnMdUnm2qhWg6jTJp5rk3RapOfapCFUiQ9aU5bZvSUcOOkBVWX/lCf2V2RJ5+84YR1ZVWKURJQGZQWqEkgZ4cIYlyrs+gJEgvzBNsFgTFSCsSUoiXSG0IxRWhJWEaFNQI84kSS4VGDKEe1Gm7S9wHD4gCoKkYtLPNjdRo0nLMkA5SRBHJILSREbVCwJckXlNFU+Ji3GkBTIg5JTvTl6p06TDidsTgZURUi5u8VBNiFvJowbLVpxjIpjRqMhYdADJNbiz2prCJXEjnO2dveZB5wTNDfWOfXSi2zdus/W/+9fEwiFkrBqU+Zzx44VlDLiobDcoWDsjM+BEJLICgIiLjxzmcZ6j+zBFuPtLeThhMbhEPHBdTprJ2HSJiBCBCG6LI6pJJiVgmEQYbBo61AOwnpYFagAawzWGor6TLN42b+QcualWlXlzBsVqMlrEq29yszvz2rmjR4EwSx83BiojLd4tELwwbWPuXDpKVqNtN6n6tCyn2D9SGBqUZUELvBP3PkCy9kjMPUogGqaFu9BOJHXoKIQPtxI+pQ+QkEkIgQOLSSl1oRRjFCSibFMhGJQOqyKEMqnbqe9OXoLiygNIgwx1qd0j8Z9FtptkrRJ1Oyhmj2acwsIJciyMbdv32R1aYELzz1N0Gpy5/ZN2nHM808/x87hgEa3C2GCaS5w4vLzNIZDjIHJtbdRlcbYEVFgCJTDEoCAJE2QYZ0gabxXmhMlKo69Aa9QOKXQziJlQNxuI+MGNjtgvPsQKR0uDinzjI6AuUZCNepz8/pHOBdw5TOfJQpCFI52M2YyjriztcNiusBGM0A4KAEpFMp4I/ipjOmTJP5TEG86aX2SWWytxVhLVhQcHPTZPegzmuQMhiOGwyFLy0v05uZoNFs0Gk1PxQ7CI1D2MZDWgz5Tr9yf5tJa0x8MZvflTM6vJCpQ5HkOzpLE0Qzo9c34UaGE87L7yugZqDqdEpeTCUGocB4XZWlhiXMXLnDjxh5YUafG+/tBKQgCdZSy6Py0fuonqzGEMkSpAKvcjOF64sQJIJpJspy1iFCiVIiUIQiL1hXD4Yhms4WqC9RKa39fSu/AN+1jnbEUec54OOTs6TN8+Rd+kX/5r/6V9w4UxpvfW+G9SaXAOW+IPxoOuH3rup8I2YAoClmYX/FWGghshU+ZFJZOq0FzZZU8azEajSmLEl1pMpchlS/aKyn9fRHbGkgtKMsMqbz8XqmwZkFYRG1ZIaWrJ9nTKdPRBiqEJM8z9vYO6vfV3//ZJCNOEqIoqomA3sw6jDxYnWUZuRVEUUikFDq3KPfjiHl+GutxibB/ixxlnlNmE1+YS7/vaqP59ne+zeLCIhsbG8jJZPY5b3c6GO2D4c5eforLz1xlPBihopC9YR9nLQvtDjfv3kFKwXg84t7du1y9cpk4Cdnd3aHVbBLFqpY9Q1kUfpqoNXv7B8zPLzAocqJmi3anhwhCVO0II6WkMppBvw8CTCBJ5zpcunqZcxcusHLiBFGnRdBszJgwU7BKGE12OEC2ExrNlMloRBDHhHGEkgrpJGmacvP6dUSoePq559BFgVIBcRj6xGklKBVoKbG2YvnkBmEQcvf+A7pnTtBdXcCFlo3uGVQYsns4ZOPMOVQQMJwM/YEfBownE9rtFtI6rt19j3u3byONpb04jxSSh9dvYScFL//KL6LSFIT8Wcrqj70el6jWulGmG5sQkjAM/F6qLXEQzCbXgRToSiNlWO//R2fRtLhSUs3CUHDeKH/m7e78P1dpXX++vB2IQqLLCmcsSvkiL89zEAFJrAiCEGO8l7Hfq46xwn7gPqjPTQKk1Sgyxv0PmDMPWI8drjiglbboLncYDEbw6BFybZ1Rt0W50uTanRvkb93l/jffxyDodeeJXcVyGFFOOty+fJlvrqbsv/Eu3YmjJQKkM0RxjLam3kuOEqmdc3/mpvGHt5p/NYBUACecDwQRht2HD3nr+1+l004oh5CokGaoGeWHtCKJqwqeunqJG7e3yXRAd75HoGKkDuhEDRphA6NSmu0m3YV5Tpy5zJ3bd7j90TtUoyGNYcT+rmZl2aHjgpd+/lXe+fgOzWUIoiHWdeg22lgK7m3n/Ot/8yF/S53j1//Wc4jIUNgCaQ/Zuv0moWwRLmzwzvuvs9BKuHbtXa4+/QLOSNAWpahNxDQ4rwKRwrG2cYbe3Dy/+8//KUsr6zgV8v1v/hHrF57i7LlLftCpBC4zWAK0hY31dZSQPIpa9Pe3CcKUD9/dRMmYuNPj47u3iFPLykKbwLWQJiBQQX3WC0SgUFGAlaAJuXr187R7Z9AGrMgxFkajHElCtxWjq9LXDvh6ZlrXPenNKGqQcHFx8VPJAvyk5cFDWzeQZiZJfvzz9sTeeezx6XcdJ78kSUKn3cY4x3PPPcfC4hJCSYzztZp1bkaGca62TqnZUM464pr1Of251hrfH1rLo+1DqMGqOI6JwnDGEgUIpFdJlWVJmqaPhTwtLi4yHo8ZjUb0+/3aY89wcHBAq9VCG0NRFERxzHg89o9pPSP8qPrf6nQ69WOSdruNlJKyKCmrciYFjeN4BsyOx+NZINL0uQCzx36QPFTzUmtbGCWmpkz1svjauv5OMQXp6gGgUAIjHFZYhsMhjSiaedP+JEFIP+31pKzf/25qVupxy4TaQqHWpU8/i2VZcu/effb29unNzeOc4IMPPqAsSqIo5OLFs+T5gPG4jzGGubkFqqri3r17nD9/fubvf/nKZR4+fMjB4SFxkjA3P09ZaPKsAOeIEx/Cq9SRDNnfw1O7L98//SRX3DUjVp99ika7RTrfpZWktGQElUZOShjmFKMxxWBMMKmww4zR8BHDu48QQKBCSCOChS5pt03rmTNwdhmFI7QerLV1+x1YjqykPv23yb/1NQtEFhJdGoT1n1ElBRhNaDTNyhIOK6ow4H6V0zyxRLYbEA8mhBNJ4CpAIkQ5Y0BSakLjCEyBshkmbXG+t0hce3TLRky4NM/Og/u0W02COGJ8OGRubInTABFBbPCqJgICBGHpEOMJARo3GNI7bGHbXVbbIUm3wXBvF5vnnN24iikyyv4B8cI8VZbhWm0qJxAiQATCez04g5OCXWGxkaIjI+Y3TiHjBs/8+i9x/83XUTceMFdAEkooCxCSPWl4oCqKwBFaiTaWXDjaQcTZ0+e48stfpnH5FGFmKG/e5WDzDtvXr5MejFB37zM/N0eQKpyGqA7upg7gdtahhCfaIQUyUIjS1ZwYP6gNVOQzd2oFcFWWGO2DI6fAKDCr8afMcr8ne8X7k18XwpMrgkASEmDCGKMrlAo4ODhkd3eP1ulTOOsDRH/SLfZHAlPLqkJPJR/1v2iNp39Y530bpunFcDQJFeIITDXS04WVE1grfQMjNKVSaCeJoxinNaW1jLSjMmDRHkgKQrpLi4QqIFKKKpCYPEdVmqAylP0x3fVFZNqiPxnRKlroUcHm/Yc8/9LLNJbmEbbk5ntvcvb0WZJuDx0pQqG5+9b3KcqCUTZhUhTYZove+jLVYJlq5wGhMAhrcVZhnCAIpWenOFsbrnvDaGsNZZEjg8B7QzpDZQvixB8scbONzjMCW4LyQQFRWRLg6O/e5eG195lkhud+4deoqEikRVcBlTFcOH0aTqwx2H9IUDp6EaBKRFUhrZeV+t5MPMZOfTKEagqgmnoK4Jw9Jss/KtKyMufWnbuMx2P6/UNOnz7FyvIKi4uL2Lk5XLtNnKTej1SKx+q42YH6F+CKPR6PKYqCrGalOudoNptINQVSPdCZpgkqCOgkCdQAnKrN9qua2VpZX6hGUUQce4bpeDIB/Ac0whd73V4XbR556b3zAJCfwNaVEv7+Pz6GlUoyHI2x5MwvLBGlKVVZEtaF7QsvvjBrR7UxhIEvDC1gTcXu7o4PXnMOYw1CBsRxWAO2AuMMzlh0qTncP+Tdt96m3z/g7NnT/MZv/AZf+/o32Ds4xDld+6uC1YATOCdx1h8gZVFw5/ZtjJHEcUSj0aTZmvPWHVKgy5zBcAjCMt9u0ust0WnPM5mMySb+vahKjQt8ZSKl99Ky1lLpasZ0nN5Avmi0MzaqENPtaWqhUMusnCXLMibZkKIYoVQdhID3tI2iGCHx6bRG+4T4IKCoSiaTDCEEYRhR1l65xn46R7pu1hQdAanCWbLxEIH1gA2wurqKUgFnz54lUCEfvPce+/v7vPzyywyHQ7a3t3m0tcXFixd56/XXeeGlF5lbXSKQimsffsT80iKlMQRCsbi8yEcffsCZM6c42N8jTiLiOCBtRH6/cBprYW//gChMwUkWFxaJGw1cECCimDCNEcozgqxUGBzjwZAP3/uAV7/8S2ysr3Li9CmanQ5hmmIC5a0uhKq3DW4pprUAAQAASURBVOeBYuvQRnPlmSvkOqMdxdy9fovGwhxWShIVIoOQ1XMnWSxXuPPhNe6+8TanX3yaLAoIgpAYxcO7d1leXaEyOUGoCNMG3Y01nj19CqsNeVWyfuEM2lS43BLF0GjPcefOPeIwotPpkOc5B/uH6LKkFSUM9g5pJw1uHj7g7LkNhoMBmx9cZ/hgh4XlZa5+9mVKWQNz8nElwF/19UlMyB9sFj+hsnnsIUcSJ7RbbfLKU1OVVARK1mGQ04LKEaiAvCxQUnqpX302hJHfX8qyAufB2WnQ3hSk8N975O1G3QwK4bxVivODu7IsqRAMhyO6vfRoDCIEn/ByEU76oZRSSFNS6DF3bj1ktD+m1Yh5tF1ysKdZ6bZ5w2gGZxoMuqBGj3AfDChf+5jtb79LEDhONOeoBnu88JkvsD3p010WZE3Fo+09tnbeY+5UxGE35t5gn//TX/t1DyIb34GJH1PGeHQ/Pwmouie+9u/2kg6kdVS55tu//2+w4x2SdkCwNM9gb4swThgWBe3mAstzG0yM4cLFkyQqYjjYp1vt0aFNIkIcMTaYIK0iMk2iTszVl9d4+uorvPO9P+LRg7s0ly7xcDTk1o3bhCJg/sQGpVtl7zBjfj5l//CAapwxHhdUKNpNxfiwz73ru6yuvsw73/qndMOY5ctf4ltv/A6LAXz88SO6i/PcvR9zev0qxk6Qcoy0gAgwNbvTVSXCQSBDfutv/0d89ff/JXvFkOVTT/HRO29z8Gib51/5DBgPUJUuo7KGfGCZZBmP9neZHCo+un+Ph7sarTPMrZss9RJWuos0k5AwtARBhgrAkSIC33zJoE3lHM89/UU63fNUxmCsoywNRVF6lhkC49GpmmBgCZTAaIuQj7c6U4BK64qDgwOWlhZ+6kqqH2dVlZ4Fb8wG7g6OdV8//C8fO4am5AtjDEop2u02KEWj3fRgp3U1W9N61ZTzllnWuRoUnNb9om6I/c+11tLv92tgs2I4HrG5uUmgApYWFzl7+gxpksxk9hZLKEPm5+dZWFiYNcgzCb2UTCYThBAz+f/W1taMGXt4eEhRFCwsLGCjiIXFRbLJhCzLWFpa8oFUuiJJvH3X9OebSmPN46rK6Z/TNCWoh3NTltPxX8Dj/y38cK8oPaPN4QNDp9YSUwDV6xGo7X9qVZ9SIBzWGUbDETg3s9L7y7CefJ5PyvunljxT0s5jX4cZCaIoCra3d3j48JG/N4TAGMu9+/cpS8PS0gIb62uMRgOkhKIo6XRaLCws0O/3uXXrFmfPnuXhw4esra1x+fJlDg8P2dnZwVrHyvIqURzVPbKpz24vPU6SBItBTe9rfjxG6vFlEFQStBSYQGGiAB3HhEkbhCQRik4QEEiJKzVuUqD7Y2RRUQwnZIMxxThjNJ6QjyeEp5aIncMKgZb+HhLOn9+yBlSrT//29VNZVkiks1QSQicJS0cVCE96UxVKZ7i79wFNY67DetRkEEoGacx83KTa3kdVBUVV4YSkoB6S2IrQVARWo1yB1Ia15hLaVMx3UoSRoCuStEk6v8gkldzY3uJi2qVhoDl2dIgwRFhpsVRkYQXCkgUVUNHqT1DDA7pJRLMoaccRSytz7N6/Cf0eiQwJm20yC2WWEbY7fidxfrCLEATdNvnSHB/cusGrZy9x6uozqLBBcnKV+ZcuMbq5SSoFZWiwQpCamEExopKWhvEepUNdMRaCXjei14jpD/cQ/UVo9pg/f5Xm1UvMf/4V8o/us3swZuetd7hw6TxJOocTAUrImuHvsSnj6bP4REmJnM7/nMDUCrBQ+YGdtpayqBDCYI2r/U/lbKB33EfV29NVCMEs/PzInqXunYUnUcROUFQGoyussbz5xpucWF8jVF4BKX7CPfdHAlOrqkIag50d3tMT2dXos5c0uqMOov7N/66mRrJCoAIvTxfGoCjJQ4UjpZGmmDwnSRuetqwrklaTSZ7TWp6nkaZIXVEECoSiGTQwDpCCia5oRRGVsyx12+xe/5DJeMgzX/g5ZKdJPikwVcXKqYsUgSLrD9m/eYvtrYecO3uW3bv30YXl8nMvkDlLhcSg0OMJQXUIQUKpPcItsJTG+sR4WXs2CumZfc77/yihCOIGTjqkLQlDSaPdIh/sIQqLdhJNSH7Qx3z4rvdhNYbG/DLjos9KsEKpDcO9IadOn6XXjYkpiYITbG7vMjE566FBmtL7zjoFwjwBlk5B0iMLgCcPttn34dOMrfHTZFMboG/tbFOVOYPDQ3ZWVzlx4gQnTpxgcXGRVq9HVCdr1jcD4GZMADGVsvwU18HBwUy6VBQFcRzTarV84nz92lUYkDYbs0OzKkuE8jKdOIkpy5IgCrFazMyQkyjyVg5hSFGWJGlAGIY0m00/dbUOTTX70FtNzV71wUDOWaT/QNTDBcn8/CLaToEWMZt4I7xnULfjC9OygiT2cJo1hv7+HnmekaYpQtRgjWD298uyYpJPKPMxo+GI7YeP2N56wObmPVZWFphbnOev/cZv8I//x3/mD3FncE567yDrmRx+WaRzGGd58PCun+Y225w+k5DETa8tsQGVrTg8PKAcT1icn6fdbtLtztPt9CiKjLIsqKrCe/oZTVlOPU09g0BrvxHi/N6glC/Op1N8KUV9XcHYiqosKMqCyXhMUZSEoQNXM8hqiYCuKpy1jCdDlPKMeu+MJAmCkCAMCcPQhw0UjiAKf6r36Z9tHR0IUgjKPMPoEok3Hg+kIAoj3nrrLbYePeLUyTNcvHiRsiyx1pImCU9dvMjp02fYPzzg53/hS+RFUTNAFb2FeXrdLjhvjZBlY06fOcV4NCKOvUSv3W7iMPUvS16UJEmLIitJ4pi42WBclSAF3VaKDOqy1ClK4bh1+zYtGfIrX/4VVBrgAg98KRHgrCOSIJ2X/FkpsBIwljKb8ODeHZZXVxCh49HDXT586x1ai4s894VXIZBoIVBpTJLGvPCZl7h78wbf/cbXOXX+AisnT+BCwcrCIh+/+Q7tXpvexiqhCtDaEinJ7vYWUSPClhXD7V0qDe3OMju3H9Hqthj0B1htSBopaZwQNVtk/SHCQj7KSLpeijUejujv7tETIQd37jN5+jJBu4kQEi19sfMz0sBPuo6BdELUQH/MJK8oioIk9Yw23yxZ9ESjAkWo6tTPGqifAghG++EA+GZXlxoxZUSrqbzTgy5VvafosiKfTJBOU1YZWT6m0o4HD3bp9eaI45Ber4E/C49YtJ/0SpxwCFchjeaDu/e4rzcYt6/w8l9b4HAnZm/nmzz/+QX++3/8Gvv3R2TtkHWVsP3u+wRZSSuViFCyPxhzeuMkqhny8Y3v8htrzzG/84j744CvffvbPH/xAiNR8l/+l/9HLly8iLG2Jvr+pA37n9Ry/tVgpzoExsJrr3+N4fgOzTTAmAaVCVmYXyQINMtrJ8n7E4LAghqxuL5EM13iRBQyyXPuPrzDhipoJku051YhlD7OyoaUNke1Q57/0q+x8egBf/iVf8PBcIdMLvL29WucP99meHuP1cUeJ599luvv/DEXLlzkbzz9c/yHfz8mndPs7xqee/oFrr32++STA04/9+soaXn5uVe4+f47PHP1VW7f+ZC9rTskssXKylKtQBO1K7zBBRX1JJRYekbnF3/l19jaesAbb7zGlUtX2D8c8vp3vofF8fRzL1EVBmEMlc3ZebTP1sMh27sj8vGQYpIhpaLb6dDuxMRpAyUDgkhCpLDS22VI2URFKaW1XL3yJbrdi2gjcKIgL4e4MqbZbGOs9uGWdjpk9N2a0T7h92h5MIXaxsPWXo5FUczYL5/G5ZwP1zDmuJWXf51H8A/8aaeMq/fO6YTHWl+/OgGNRkpoIvIiZzyeIIWc+UZPa9OphFk4ULPG1g9mwjpYZHHR+1Nubz/iG9/6Ns1WiwvnL9AfDrh+6yZz3R5z3S6NNPUWXDWwfZwBOgV6jTF0Op1ZvzJVn43GY4Ig4J//839Omqb87b/9t2f7WVr/XF8PlzMpf57ns55HIuoQUnsUrAuz/ui4kgEeB0+Pg6rOef9TrKMu5TESrPQp1j5ZvL5eeBaW3/vrAFwEtjJkkxGBcDRa7RlLs6jzHD7trNQnAdSjx38w7Pj4909v2arSvPHGGwAY42g0GgRBwMFhn83NB4Cg2+nRbne5ffdtlpaXGfTHVFWJlIKNjQ36/T77+/u02222t7epqoqNjQ16vR7D4Yit7UcEgR8aNBoN0jT1nsCBIowUTrjHTrOf9IpbKdA1iFRmBZWV6MqrQcI09l8X/oPk0gDRjIhXekgHKYKuAbTBZBV6OCbsNQkr3ztVNTklsIADLWuW6p/D8/53YWltUc5ShZLASiILReR7TgKDrCZkD+9jA0l5eEBzYZGqFdNtdLDjijKI6c7NM9jdgUwipSZSlkiUCOOQTqJLizQFSu8SiYory8tIFXN3NCJMExrtNrkztNIunaUVDvWE27sP2Y8kbScR1tKRkl4uGIeOQeJItURJQ+BK5CTHjYeoJMJmHdrtDhExwaM+YTQiW4gZJQPSTs+zk6n3diWglXL686/wB7ubbPzNX0GeO8HYSarI20lJZ8gjSU8KOkIysoaxglAGRM5yWNv7tAJFt9A0Nx8h7vQoF+YRVtAnRRuB6M4z/9Ii7YMR/5//1/+DG3NNXvrV32Dl4mVvs4DPsTGlxlEPVqwDaZEi9H6qQiBRaO39i531nt5TjGo8Hte2L0e2W9MgWVXb0PhcJq9GyPNiptiYnvNSCSSSQEhEFFNYA1Jx/+49th9tc3JjzfuC/YTrR6oe8qJAHjvspmuaCjudMj12wMgjY/fjITLa+CT4khKc9VKd0DLf6jDe30FFEd2FDofViDyf0Gq2iYOUqjREjaM0RKsE1CFCK2cuQJoSFyPufu+7RGGDp3/pN5nIkBYQqIi8zIlsydbNe4RJl86ps4SLa4zHA+Y21lhbXMboCjfOScKIotODi5fZ+fgtGhKqskBXgiRQOFTtkyZRymF1hVQBUsGoKJFBTOA0GBAqJAoUNCSDpM1wkkFZkaYVuTZkew8QsYS4yWQyxjnIc8PDO3dJGhGGilEuIHR0Oh10GLO9dR9RFKzIlJAKi8YSIl2BqptDgzdB98i0nfU2RwdbPWm2FluHV9kphFUXV1Vl2D04ZDyesLd/wEH/kNF4yMnJSdaqdZqtNlEYHtOrTIsO+VjB8dNao+GY+fl50iQF50iThCgIGNceiz7MQPhUTSk9QBp5o+PDw0MqY7ys3Namy6FP8y7yCiEF7VYLMRrhajp6kWvyzCCcwRmBIADrG3ClvKzHt1u1r5Lx18gJWadPCy9FE6DCCKMr4jghDAJG4wMPDhpLGZZIIcmzDCVDkrhJHDUQIvA2Bc5RFDmVrjBG42xFHElyBatrqxjreP+jj8kKsC7iN//a3+Dhw13++JtfoywLhLDep1R6IFiomED6+wJrMGXFvbu3abd7pI0Gq6sniOOUMA6RwlFkltF4hNYF46xNu92i2UhptTv+XqwlacZYDBqtNZEKCcIEFTRqG4YApTxIrYLQM4WV8Kn1rkLnFcNBn0meoY3x0t4wQhiHdgbjDFJC3Eyxwn9eqzq5UQUhCElVlrSbTcB/drEOicToTycz1a/6vhX+V5HlOOsQ0rNUnXVoJ3j2uad57tkrPNh8xPe+9y067TmuXr7Co80HbD96RGe+R2dxEW0to9GITqvNcDSg39+n3UwYDgbeY9KMieOYZjv0jWYtyfITcIm0ksj5oUsUxRAoJrpCxgm9xQVEEPo+zcFwMmKSTTi9scF8r+cZ/ML6CaoDgfD+wuIJRnv9Z2O0/4TonFu373NibY0v/Povc+fax+zfvMni2ROIJAancEJigPmzZ0m6c1TDCW9/9Zs8++ILRAtdLn72Obau32b3+m1G+YRKOD66do1XP/d5luZXGQ6HDIc5F566grWCVEVU0hIsKIqiYPdgn1acMB6O2O0fUGQZ/b195k9tAIKH9+8zzEbk4zHZ1j55f0iz00I67w/0syr3z2n5QsM3qVYikWjriMMQnEBWMVblEFVYGyKNIAgsMkz8PaILz74vHUqCDsE6TVJZcNoHhUQRzlhSJQkovbNpaFAKbBzQiFP+ze9/hT/4w68yv7RAlKScOXOap0LFYb+BNisEwo9vjo7AJ28Ah8DgBOzs9+nvTnjUB7f8AnEv4tdeOc3eBzd5uHmd4ubHuFsjdOA4cCFpEJJVGe1uytZgxOKlp4naJbde/2948ak5rn+9jWi2KUtNcH+f9CJ0mglffv4VpnL+IwuC6VjxSXbp8fXDbt7p408Cp5/m/fTPttwP/Mdx1q0DNA/ufcj3v/8/EVKQhA3isEkYCaLQEqiYJGwyrMZsbe8RNxXtJUlnrUte+UyAtZMXcbri4GCXhw/v0JmbZ2X1JCpokIjYS+9FwOL6WX7rP/xf8e7Xv8qNj99godfl9sMJKg54mI141De8+Opv8ce/9z+wuHqb7uoy81zE5Iad/G12dne58PyXSOZWSKOItEpQ5x337twlFBHNpMHd2x/S7aXEQVozpy3YCmqGyRQxcqZCW8H8whq/9mt/nd/5V79DoiQrqwvsj3b57rd/jwvnrxIEKcZq+oMh40lOpkucKwjCgFgpeknKUqdHEkEQgVINlNUoEYBMCKMGQsWcPHWRheWn0CZCuwlFWSBFSpCGvm6vdbnHP25TsNHViiGPB3pG+dTTLYyCuuYuaNa1wadtOQdFUR0bzhwH844+a9M0gyfX7FP6hFx8yrTX2vuilroizwsqbQiVHzof/ZCjvzeVsLtpGJUUxElCqKZKtYhQBJSTgslwzHPPPMtct0uZe3VApSuCIJj5lCopCaIIVxM5RqMRUVwTG4IA6xyD0RCtNdpounNz3L9/j8nEkwo+97nPzfxPp430lLk0fezJkC5tfbfjwIegcAzkm77OYwnzj78fj78PnnTqUCrAt+34+mZq5l1fPwtI5Xd+D/g7qrJkMhkTKEXaaNTp1B5M3d/fZ2FhYWaL8GkBVX/4EK5GjZ1H+bx52eOs3x/wU3WWre1H3L9/l/X1E3UQrw+s0mXl7zVX0m5FNJsNNu9vI5GcP3+ezfsP2N3bY1lI1lZWGI8naGNI2inD0YiPr33M6Y0NVuYXKOZ7FEVBfzhkkE88Uy2MWFla8vWZdX6A80MtMn60JerL4C0pNJU1JEAgBFpAYAwuK7Ghz4YgCBBKIaXfmxCSSkls4ED6fijoJIhQYYSbDYqmVhFT7osVtSvLzxZSaqxtYs0+O1t3MNoDz7FIQIRYadHOElWasCxwVUl3fo7GXI89KRg1Y0aRYowl3Vigv7ND24LUEYFwFMLVIdwSVxjk7iHZaEiyusrFziIrIsYIwf29LUaJominiD1DIHpEF0/z6NFd7OEQTEI6iXDNkM6wpKUNjoA8gFwaGlIiCg0P+zT3C9RBgegPkXqMZAOdKKqq9FJ4C075cEkVh3z5r//PuHTpMk999gWKQNEaCfqRosotyoFxBakJMTLgnskQTpA6xTiy6MqSqAiDJDWG5njA+KN3uHbvPkvPXOXCM8+jwhbluGKcFQzvPaDY2ub6rT7X7zzgy//+f8C5l17ECP98jBIEDqzRvh53EhlKqEpkNkY3QzSQ2BQhI5QAEZSUFtBH3thBIGdg6lSZMQ2bUoElFiFKOSaT3HM8jbdkFE4ilUWGikIGGBKcnaA03Lxxk8W1FZQSRM7+aIDoD1k/0t81xmB4/BCfFuTwg/yEJ32Jjj9uhCAATP23jAChHFEYMxmPmWQTDvf3yCZjGs2WP0jq1O3KWNJ6FluZCgJFS8RMxjm9OGTzu9/BqJCzX/oSOg1IKsn2/S2/MQ/3KCcTzj/3PDpJUUJhM0t7fhFR9Ll162Me3N/k/KnzvPSZz3DhzCnu9lJEOeHB9Q9JohDrHKUxYAvCGIR1GOuBMougKjWhEBRZ5pODpUBrz4aL45ik0SQbp4TOoKQlaUV1QrDyDtLjCYO7t2A05v69La68+DyHB3u4qoVopjSajk47JRRr7D/Q7FZjeiFEANZ7GzkZ4pxBOI1wpmaaTr0nH/dLrYmSs/fLzmQZ/v3y02HIy4rd/X3GkxGj0ZBRzQpcXlomSZLH3t/pzf4XIZnqtNvEUYgQ0SwBdDKZeM/eqZk+kslojFSKoiqRSjEej5hMxj7QqdmEKZO60jgLxvr05jiOKYqcstKEYchgMCJpeGal0cYzOeTxBNXap6uWjyiluHf/AWEUc/LU6RmnQAjP1MuzjKhOtnPOYoxGVyVGV3TaHZyxNJoNdDRlIftCeDQekyQxaSMBYsAyGhzQm5tHCIUKE/69v/lbXHzqEu12G2sN/+t/+A85PDzgrXfeorI+jAFbT+mnE/XZ4S2oipKPr31IkjYJAh9IlUYJMgwRNkEJS1UV7O3tMR4OaLWbtNttmo1WbZUQ1RMkUctrBCqQXj4h5cyb0FoL1jMgi6oCW1EWE8ajIXt7ewRROAPxFb5oKXRBqTOieMoWdljnLRKUkJiyqt+/aCYBMMb4Yr2ehn0q1/GitR4imTpd3OKLL6kkBLV5tnWsnFhhaX2FQEVY41g5tc766fUZY+LmzWsoKVlc6nLY32FxuUdZjUka3uMUE8/YOszu4aPGFOEYjjL6hxOW1lYpnUNFMfMrK6AUth7YjEZD8jzj1MmTMwNx8MOMx4pq6wc4VvhhDs4Xhbb2zlldWcKUOS9cuooNA6wznL36FHubDzncfMjcxjoiidGlJohjIhfSWFnFLWoyU/L2O28z12xz9tJFzly9xOH+Ac1+n4PdXS4/dZFmp8nWgwc0Gi3mlpb5+je+wWdf+iwH4z47u7ssLC0ShgHNVotxf8D21hZ5ljHuj5BhSHdxkWw8YW9nx78X1uLKkoOdXZobKyAUklrA8enohf4Sr5qlXTc8QjiSKCBUkkoXBFIhlGf7qzBCmgYRIVLUTCRbIl2OMoJG0iJUEDUCTBghkoCOFmilEGFAbAxaersJYcAIr0rRtsJEjjgOODw8IExTFtIG/f4h/f4ha0vzqCeYcJ/8ShRYiYws3339Td65vcmd7AAtIM1Kbnz8XeRgQBz1OLm2zIdbEywhhQjJY0GVKh7ZMaUU3L52A70UgLZsfv0eX5ncx8hv0Q8UKQWT27exj0YcnBiwtLzoh3jGHIEFM/3vn+UG/STw9Rio+pf9Xv8T8GRv+wA4x2R0yNe+8q8IKel2WigiojAgCDVOGsrSMhrukesRaTtlZe0ECwtn6HXXuHfvPqdOnsfaioPDB8hQ0ui0EcLx2mvfYH5xkeXVE7S6SwShBAKUbPP8l36F7vIi4R//Lg1t+Hg7Q2Sa6298nyav8uov/hZlOSZQC3Rjwajc4u2PrrNy5lkW166QpClSVDhr6HZ75Mt9wlDzcPM+J06d5I//6Kv8yq/+qvebthIhdM309MsP4Kd+jlBVjl/99d+gv7fN69/7Y3b2ttg4c56Dgz7vvf2H/NIvfpGz505y+94OgdEYFWFsRZhIGs2EKBJEoUMFEWVhaLRDhFCIMEJFAfMLpzh37mUq49AuoygrnJWEdRrwtEaZDgfcMVb4FBvxHtxyRvTQWh+FIFlLVel/u/fTj7Gmr2EKBj7ZTx0H9Z78wM08kJ/ov47/PX/2+lu9KkuyPEdISW9ujjzPj/xBn3xODmpHWryE0zAelkSB9ygNgoAojPjsK5/h/PkLpM0GzjkODg5wzhEGdQhUbat1//59Wq0WjWaT4WDAYDBgbX19FgQlhPfALivPjm532qysrtI/OOQ3f/M3Z+8pHBF82u22V5LVgG2WZWxubnL69OmjOvOJ6/hJDMonAcxPDO0TYma9ADWAb21NLpn2zUfeq7JG3LLJpLalikjS1PeSx37+ZDKh1WrN6qe/SED1R1YxuBpGfcIn9UkrBThiIO/u7tSfz8drwzgK+ZUvf5kbN+6wtrZGFCkajZTbt29TFAWnT59mOOmzu72Dc47VlVWs87aEURRTlgXX79wmTVLmuj16c3N0V9tArXYLA++VK49b8fxk13em6yt8bxk3UgLpw9m09CSEpBFjpKPEUYoaJ3EW5WqijwhxElRNbEKBL86ZkZ2EPBqc/AxEfWIpi3MSu7PP7/3Tf8TJl36OlMTngsYJMklQkSKc5ARGI1yJ2y5QwzGt+R5yrkFRlYh2iFrt0T/YJQ5bjAYTKm1IIkEoBRgfb5g4QWih3N7jYGfA6sYGkZCkaQPdkuRzEcVWyXxnAdVpM7yeg5AMLq4zubtPNB6xVEriwpI3FQfKIsLAhynKgMSGhJMSN95mdLDLeO8h0r1EEDcpV8cEXYWRsiaoSIx1zC8ss9RbpAgcUluUk0SFwwxLtBAEVUU7SjkwFfsYWiLFadgNDRWOhoowwtESsFAZ+jt9xnv3eXT/LjIvOPPsM8g4QeQ5uzeuYw4GhEiqUU6oDWiD1gaCEIUgFKK2+ZMoCyY0DD66xfj2LRZfuoztzVE444kM2hIIhdSGqA589b7a+jEg1Rj3mIe1c5YkjZFS+ZwB4e1IrRNIpfBOtR5zlKGkKiwffPQhz3/+c6RxMBuq/bjrRwJTpwfPD3jG8OSB7tf08U8CXnXtsTkNTbICUuXBlCr3cnxhLK1WCxVEVGWFioyXgkyLJOcIlCSOm1T9Ic1Rn7v3PiZpNZhbP4sMYsbZEDvOCcuMyWjCibUTqHMX2EcQVJp2pGimksHuNiLPWV9cIgkiur0eUlrSKGR1dZXuL/wy30GydfMDGoGCMEIXGWU2JIlTkCHIEO0EShswFU5FVFXuKc61jD4KI9qdNoODXYR1OEwtZw2wpSQMFNlwl91yxHbUYOHERfJJyXgwJBAQRhFUlpYQdOKUcP0MD7c30cUhy0oQM6QQyqP51iAw4PyN5C2Njoqi44DqUVFx1BBNg8Scq2U3zmIsjMcZt+/cYzSaMByMWFtbo9Pp1Anz/p0OgoA4igij8BPljf8219LSEqPRcDaxkNKHFAF1IJHAGUdRlggpyYuCvMjZ3t4GoNVsHQM48SBlnNaWAWF9LZ1vRmRAWVToyqeh7u8f4qStg0ymaLTw0L/wgLkDVlbXCaMIUafRT/89a/zGkWUTOp0OSimSJOLm5j2UlGSTMadOnMDW8u68KLA2o6q8DKrZbGKdQUrBaDQibrQJVYh1sJp2OHnqzIyhYbSl02nxD//hP+C//r//V1y/+THW1aEkSuKtGqR/jsIBFoFgMhny8bX3iKMQJWBpYYUkjBFBiCTxwKTRGFNxeHDAaDig1WrSSFM/TIhj4jgkUMrfH0ZQlTlWHbFrTT3RVVLi8Jtolk18KregLtYVuir99QvBWY1SPu0vzxVCSIbDEcI5wkiS5cUsEXYymdTeK94rqagl75/qJQQ+B9Z7whknkZ5mA07gjPbm6nhwVUgwwrPshHM46f82OE6dWffXyuTML3Q889QYwHiAhQDv9+sLPF/kenC6Mn6odTge0erNY4QkTGLml5cRQYCuG62qDnY4ffrU7HMHjxfks4GDOApME47ZPp9nE0IJwgru3buHXBX0lpcwSmHTmO7aMh9/5w2KScHGlYs46e0xhuMhnbkugZKcuniB7twc+dY+X//dP+Dsi0+zcOoE3eUFpBScn5tjmGU8vP+Q0WBMd2GRJE744IMPWD91kma7BdZSjDP2+ge+IO90uPNwh/HhgLn1FcZlzoO79yjGE+aCCHTF4d4eo8PDupl9HBP/2frk9Xiz9sPN4B9rJoVFKksoodIlYTPFiQIrHEYrdFZi5Yjd0UO2Htzl4NF9VCiZO3GStaUVluabmEqzHLbp2jmIEmLV8IMkKZDCoAwEVmFlSCWtZ8pLR6fTotNucGJjldNnz3D61AkuXTjP6RMbj3nkPnnPT5cVDlAc7Ge8t7vN7t6ASBtAM9qecFdK7OFJzpkmv/Xsc/w/7/8zRi4njAQXlta5e+8+zoZ0rSNyGdUjSx5WhLQZxF2ktvTmUn7uF59jdc9xUTTZ33rE8srSbNB9BKTO3oUf9u7ww5vMJ1mpf8K3/iVcR+8ks8+yww+mta74xte+QjHZp9dNiQNqRYUnCGirmWQTJkONdYKFpSVWV9Zot1oMB0N+73d/n7/7d/+XrK+dZHVlncl4wubmA8ZZwcLKGaJAce3DazQa90kbLVbXN4jiBiKe59RTL7K2scpbf/AvkG98xCBcYZBbhoMhF6++QndhGT3Y5c3v/RP29gqWNz7D1Rc/R9RoIWkiXAlBBRYWl5YZDgacPHOGO7dvsnFijXt3r3Pi9FlMnXQvhfuBWt8cY/RZa2k2O/zir/51bt+7zp1b17n2/vssL63x2nffQgr4tZ97iWs37vP2ezewpiBphsTtABU5pAwIVUKzk+BchpOKKI5IGnNcuvR5ijxBk1OZAkFCGEYYMwYXHlkkHXtuj+0nrh6wSA/OTeWC/tz3NZHWf/Fg6if1UtY6iqKcff14INEnAWtPjkV+gA9/rC+zdU8wyTLPEA0Dojiu2a41YGOn6fOPg46i/oeCIEBXFYf9ATs7O4D3cF9dXSUKQxbn5/2gVwiilcj3G8eewxRg3Nvb4+DgAPCZBLs7O8zPz1NpPSM57OztImo//eXFJZYXFh8D66qy9LZcaQrgPftrj9QgCNjY2Jjds0eEhx8Omn6SrP/4149fU2PMzIv76Mof65Vrtd70e0ejEc65WSbDkVxV1e+7fcw/9i/Dmg7d3REhtX78BwHVKZAqpSTLstl7oJQC50OAgzDg0uWLfO7VV5mb6zKZDFhaWsDojM3NexweHrK2scr5ixfY393n3uYDlpeXWV5eJs9ziqqi2e1iKs1okjHoDwmVIo4i0jRleW0VISRWgFGgjPhzO7oCoYjCiFgoFHVg1CSnDC37RuOU349kFJOmntXtBJTWoJxEWb+naecHRZLHrSV+ynEkf6mWtRpjS97+Z/+Kw2vXWP/cF7DOEEjn/WvDCNWMMdkBmatwStAxhkYfVFkSVR2KUCAbHr85zHI2Vk4iYs3+9jaJFChX0Qw8S19J6/MsFChRsf3oNu0wYGG+x5cuXmXTGT7eOsCqhLQUrGcRbMyhz67y7uYmp7stAhkwPDhAS42w0MoEgfAszsoZdArKGQKnCfb2MB/forV6CjHJYX4OVIASCmMs1kFhBYEIfW8lNHlkUXmFO5xgsTSEohs1eDA6oAojGtqr+yprsFIhnCSUhoazzFUQI1hzisHOPt//yh+xv73DytoKcRzy6OY1AiuQLsZZCEOQyiGr2shXCJz0d7F0gLBUbsKD73yf8t1r5JM+q7/6Rao4QEQpWEfoJEo4RChxzvieX+t638D3qJQzrGdrawsVCFZWlonimEbDkmVFbWHj/CfIQMtBfnBAEAqIA0ZFzmgwojHf5SeFq34ik6BPmjYdP+yPPy5rOYS1Hhk2zngBtHQ0Wl5ek0QRwlqCTofxaOAPEueTdJM4wlhLnmcEgWegpa0WO3eukw8PUa02rrtAsrTK+uIq9x/e59Ht64STAc+8/HmC7gLb4xwrHXEYMSpyzPCAS6dPs7y8QV5O+N6bb9AfHvLuu2/z9KVLJGmDiXacf/ZFJof7FPtbOOWDL8qyICsKhDQEkUMKySQvaMQRRhvKwrP9jNVMKw8hJY1Gk2x0gBIKZSVlWWGLirjTQAUam4+xUlKIilE+JpzEqCggTDMvYyZFBRFBI6K3vMpwy7AzGbIQRDgsAQaLwTiBtgJt8cXLsTCq6fv0+Hs3/e+pdOA4K016xpmFMivYfPCIPC/Y2zug1Wp6eUt9CimliOOEKAof8yL6aSxrvZRnOtEOAp/41ul0Zt5HSSMhyzLf8Ar/fFdWVsjznCROUSr0RvU6xxhNnk8AgbEGrKulT37Koo0jzwtarTb7+4eUVUHgpoFKR8nQSH/9pJS02m2EDGbXXNTX2DpLEBwlSkspkEHAxx9/zOrKEutryz4lV0BZlYzGE+I4pttr00hr6bpUTCYZSoWEUQzOTzelEDOmazYZsbuzzcOHD9g4eYpf/7Vf4/b/+xa6BrJmmc7OgpPU81GEdYhAcHiwy7WP3iOQoJxjaWGZKIxQIsBKgYhCrI0w2nuljoZ9JqMBYRgQRSGR9KFeftouwHmgVuuKadCL1iUCCMLAs76do9Fs1CCrwFQF49GIuV6Hqsq91CENmWQ5h3lBHCdobUjjGK19kZvnmZcQQX19vbdWURQ/FLj5i17HB1cA2ljUlP0hBaNhH2sFodD+/cF6L1WhfCgdDiGZsXGFkARhWEv3PbNE6wolpibggsoxAzed857L1jqMA2tgMsmI0zZaBYQqYHl1DRVFVMbU97I/+Obn573H9GMX9/g+M21gjpjw0xYPYxkPB/TaTZwzdHttxvmE7O4mvfVFgkaMiiNOXr7ArQ+u0e71aK0sIoRjfNjHlCULq0sgYG5pkUmS8LnlJcajMQJBUVU0em3vAJtXtFotkrTJW997nZ//hV/kzv17BEFA0khpyJDd/g5rq6vs7+2xefsu23fve6/lOGLz0SNG/SHSOMJQosuCssxxxhxzLBD8Za5+P4mV8hfd4Dnn0LYkz8aEMiSKUh+6p7wXc384ohc4btx+ja9895vcu3Gfna1HBBfOcf5zX+SXm4sMDrbIyz534hayFdFdmuPCxklWox4yStDKN0VWSrQsUNbhK2bHqVMnefGl57h0+TI//6Uvsry8RCSD2nvJ/ACAXt/1sz8hHE74gMBShqSNmLe/+zbNy6fIw5RgP+fls1/gwXde5wufP01XBoyzjMLmfHz9Bi5Nse0u0VwP6w64vNrg/EIfm+3x1o0H7O+GHD6Q/M79Hf53f/PvYJSkf3iAMfV0n58U5P9BEHXKc/vUMv3/rMvV/yemtdM0UEZiTcmbr32D69feZnW+iRQloZoGJ4LWhklWUOQlWjt68yucO3eRldUV4jQhjCL+8//8f08QxlinEDKk1WlyrrXI7u422ajP3v4uYdgkkAF3b11neLjNaDTi7KmnUc2EsDvPqed/ERm0uPFgi2DuBGevXKXXCXh483s8evAGNmhw/pnPsrJ+kW67g3EOLQwCQSBjCDKUa7J28hS3rn/ExvopiqrP9Zub9BYWCMMIZwIstbfwVNpc141H/mQWpKSoDKvrJ2m3OizNr/K1r3yNMhsTSMvh7h6hCFEmo91u0pnvkDQdUQiRSD1jWmSEKoIwRcVdLl/5ItY1ccJRlQ5HQBiC0QW4oH5ffrCmnT0nmCW2T33Ep57J09fhFUufnnV8v52m00/r1x/Gopzuxo76OnzC/nwcdJ6+d9OwvSiJvRWV82FIQgk/nFfyWFK4Z/AZY5HC++xPJhO2t7a4f/8+ZVFSlgVf/8Y3eP7553n66lUW5ue9L6oQhFLhpJyFR9n6uczNzXnVURDUBIicra0t/15Zy/bWFgjB6toaZVnOhq5TFqo2Bl1VxHE8q/2n3qjHfx3VQYLJZEJRFHS73cceP6q/5Q/UX8ev+5N97nE13vTx4z9DSoVzZhaSK6Wc+aFO7QyOq/mqyts6fOoH/Z+4fpApfXxN34+pHYMQgjAMyfOcXq83A02EgCwfk+QxQrRAaJaW5xgO9lBK4Zzg/p177O3uMze/QG9hgSzPuXf3PkuLi8x3un5P0lXN6LQI5+/dRpqi4ggnJA6HtPAklPpJw40fdR1MRgyyMZ12GykEaRARSIlqJqRhiB5nGCXBWKrhBJQiShICJZHGIYW3iNDCexM/dnU/rQ3Lp2QZZ5iMB7z/+39AO8tAeYKYdRaZJJRhTDTXYzI4IJSAsGRoZDkmNIbAOaI4IG3HVHsTTrV6BGmTrdEettOmvTDH5q0P6YaQ42iiaAlFKKzvMcuCYTVkaIZ0iiHPXHmGy7/5JW7c36e4tcNu7HNbursF6zalub6OWp3n+rtv0sDSzksaoxwdgMXgIsEgsghrSY2kYyXm8JBWVbA/GhC4VSLErAd0gcQIhbMKZy0FoGNLOJhAkaGsZqXVwUrFga68ylMptNO0pERqgXSWUFpSKVDO4QqLC3w/PelnfPD6e1wP3wNpyMYZTghP3kkjvv/BO/z1V14kCEKqylHhqGpSWSwckZQI5+0ROjsjdr/3Dr1L5+le7KCNx1mUE1ghEUr6oZ45Hizla2yPsQiktEwmE8aTEYuLi0RRTLPZQgjFcDDECkFVVEQqgDzn3mtvkDRCTjz/DJlUvPaN7/Dv/cavIn/C6JQfCUydHjJPHhaPfZ3HS+upzGQGok4PGOtm57wVHq9pNJuoIKhRZ0ue5b7xVwoV+JT0LMtAKFQNZimlaLVDGnGCEwGVqlhod+ksrHPY7/PwrdfIRiOe/uKXkL05DsoJtqjN6M2YxbkWV5/+LCqKsc4RBQlPX3qK9z7+GOsc9zcfcOLMOdJWk9VTJ+n88q/xtX/9L8j6u8g4RoUJIrBURU4xGSCEJIxSNIH3dhACq3Utvw3Q2l+/OImZDP0FCBGIOMJGATpyBFGMrBw5IOIIGUVoqyi0I8vGRM5RqhgtLApLL20QL6+z+XCTMrcsJY6myrBCUsiA0mqczhG2PpykfOJ9kzVdespUOf7+efNsD8QJrBU16Ocw2ntg4WA0Gs0K1+PF67RY/Wmu4bAPeK9FYyrK0k/Op0W01pqwHTGaTCiLAuccaSNFSUWj0UDVHry6qui0PbNzPB7jhKvBWVX7zHomn8RycNBnYWGBOI4YDjNgOklWHousD+j+YABIlldWPGO37manBEMvDXV1CqqegdFlmZM2EjrdDkIKP5lXil6vixCStJ7Cg/f9klISRpHf3PAJdUZrimLC1sNNvva1r9JsJHTabTZxfP6zn+W3f/sfs7W/g2BaTFqmUO/0uQahqFPvLXu7j/joI1BYnClYmF+ikTRm4JkUCuIAoyOqqsBoTVVUlFkGxpvqxzUDYtb4GM+qnYYYeKDVy3BUEOBwqEChhMA5QygceZmxf7BDs5UQhgGJjRlned0cuRnz0d+fDmv9e1hVFVJKiiKf3RefxvUYi6RufpABlbEgJK3ePLqsMPkQ7TSBmA5HzJFdA8Izd4Wi0nXwjPEon9Ea4abNjaSqDEZOJf7GM9ydx4eKvCLPNcYKkkab9uIKC4uLWFztP3bEVmo00seA1BmT5Qn/LP9nZnQaUfvDVmVJM0kxVUUUSKIoYm39FA/ubjI4OKBne6SNhLmTa4RJzK33rrHqDAsrS6yvrPDWd79PK4oRcUij0SBptRjJMd1eh539XfLJkI0T6wwPh8x1u3zve6/xhZ//IlVRMhyOmJtfoKwq4jTh0a17jLIJ8+srUBq27mxysLPP6skN7h/sUxUlo/6AuTTGVqVPWvVxfjMGznGuys/Wn9MSPvUzL3KqwhKrlM0HD7h55zr7g23KbMgXX1zmzkd/yK2373NwGCLnT7J85VXOvPCLPNu8SPM719leXePj04vs9aDn4HA84WlnuBjOk4QpQZSggACByKGU3qbkzNnT/Bf/xX9GknqbGx9cZXzh9ycQPKdHorACaQ3bDx/QHcH+7gG5mUBDsLbR4ZUXr/LGt7/N7177Fq+V9+mrEhc2uPTiS8SdFrYlOPeZz7B84QWcMDzbrQi2v8ZGdIu/Md7ne98/5L/7xx+yuyXI1lP+0fe/ylPnTvNKzaqcqg+OPoB/2jrm/zd9MU++uH+Hlp+BHFPrTF+7BZzhxodv8eG732Ku49n/QRAgBYRByN7+AaUp/CCdkHanRbPTIIwDgihCoFAiRhGBU6AEQnl53s07d/mv/6v/G//X/8v/mdPnLnD35nVuXPsQQUCRO6yR7O7cZu/9XdZPniGZX+fEsy8SmG9w7/773LkxYbjX4/DhNirtcers51k6sUajLXG2REqLowSncMQI1QI3JoparK+fYW/vHtmgYG6uzd17N7h4/jkKU87YiXDEJjwulTbGYIRDYrClQ+uIMO5x+vQphoe7DEcT1jZOsnn7Fq++eJm7+weINCQMvE9ZKENEoHCyQqoQFza5+uwXiNNV8kJT6AlSBigZUOmsDvaJ8aGIn8wwnPYs3p6pmIFYURSjVG3lVNetQfDpA62mQaqfJPF/cknEVAL4GNB6/O9MfUSnfy6Kwl+56XBleh3NEdjlaum5D6MKEFJgjaWoSnb3dtnc3OQ73/kuzWbDK9Xm5hAPH/D6W28yysZ85oWXmOv1vJpLqVqtIWcMUWMMURyzsLjIZDKhLH2QpXGOu/fvEYUR//p3fodOp8P/n70/j7U1O8/8sN8avmHPe595uufOt+rWTLJIFkmJo1otSmrZ7rSFpJ12uuNGAgdwhr8aCBAYCYJ0ghiGjSAOkCAI7Bgdt93uwa3uVqsnDaREiYPImuvWcOuO59wz7/kb1pA/1rf3OffWJVUUJarY5iJu8ewz7P0N61vrfZ/3eZ/nV37lV+aA8Awspfp6FjefBczhNNY4C37O5u2jYNnj2Kg/qL39UfA+XMpTKYmzefNkMmEyGQHM2aizc5gBqWfjojwP3VRnTdE+yiDaB6/TBxniZ88PTpntWmm0VAz6fVZWVhgOhxhj5zIcJyeHLCy2iRNJvV4nShKsBSVj2rUG48pHwSFodrr4wjDsDzB5KELEaUq93QhtvwRZCpnEOFVp17oAWP5xd7Cz82s2Yq0xONAKWRjygwd0L2xRCrDv77H7xjssfOZZaquLqNIRycCwCxJzHu0lyACmzljNPx0fbiQ65sHJHv1791larKFKAzom1hGJrjESmsbaBsOdA1qZJC1LMmnJGxLlHLHJ0a6kUZbElKg0IROWkZ0GaYh2k0npaNRbTCPJZDhAuRJdKkTk0U4gDLhyijWOt7/1u/i7i1zavk52YZkyKUFrWic5+ajA1lOOhhOKUUH7iS2ElBy/cQtiT5aVNAtLikKXijRO8DLCFCP2776Pu3QRW+ZYpYnRCOWxOITUOARYj1SaLPWY4SF6eEykJZ1Wm8moZCo9dQTEjtIULHpBasB4RyIszTTBKMVRbnjgBBkxvqxAWl8wKUcgJBpNGZXIpEW6sICu1fHWkmU5hQChIZISa4vQdu8ET33iRW68s4OykqaLqDlBf5whvMJKjcUihEXrQOpydtZR7QGHlHq+T3U6HY6PD8myDKUipFQ0Gg2KPKfIcoSE0oEzBSfv36Icj1jdPofu9Djc2cNPclTnRzOg/FB/PQPGwkYo5m3732/T9v7UGc9TCZbPKnhChNZTQUj2o5i0UcchaPW6nBwdwgSkCoY0SkdorbFlyXAyYHFpGS0D4zARESqO8UJhpWdgJ3DvFrduvInNTnjuKz+H7SzxoD/BSg8S6tmU5y6d5+KT22TOkGhFnmdEkWKx12Z9ZYkb791CRzUODg9ZWOyQlZ7u8jpPffyzfPdr/5QsN0RRcBiPoqotlgBiSRWYX0WeIUjwQFHkRNV1rNcbHBIujseC0nihMMLjjUHlhnq3Qa+zTBq3sEJhnKfMSzJyhtGEVAvqQqCEpNFosLR5joO7OxzlRywsL1GTHmMcw9GI46OCInPzqvPZexY28NPAYnYLz7aqBDaxr0DxGYtMVIzVU62iOU+lqhTPNsof5/B44jTBVkLzUWWGMx5PaDQaZFlBlgXXVqUUJ/2T4DrnIUmSoKtRVT/iJEFFEXmRk1WtbkpJJpMxUZRibUHkDYiC8XhMr9djOBzMmbHhfez8WWi2OpTGBEC7EjoPQV1otXTWhIqMUvMFo8gzvvjFL9NopBUoE1g/cZxijKFer8/Pu8hDEBnHScUMCu3SRZEzHJ5wfLjPG6+9yivf+w5pHPELf/7PgzOMRn0+9alP8t/9o380Z4GeHULIoJpPcCv13mKdYf/Bfd5whiKfcG7rHKtL6zQaDdIkDawFoZACtFQQO5y1lEUeNGitwRj5kBxDksaoCiiO4xg9NzIK1ykkqgolochyynzKrd07xLGm0a4HULSSSjDGBWOJGeNSKZIkYTqdYEw4P+ccw+Fw/vVPwpjpGnsP3gTITumYpNMlm47I8wzlQ1s+VWFKVgG99QW4ACp7Z/DO4W1lFOA8FcaCc0UF8odExTuBR5HUmuQ2bHTr5y4EQMf7uR6zh6pVTZKmSZVMgD/T8/XY5ERQuesG4N8VBWWWoaXElAU6TegtLDAYnLC8vYGd5hw82CNu1WkvdKkvdFm/tE02nHBwdEh3bYnnP/tJxvcPuHPnLr2NVdYvnmdalvRHIyb399m6tMVoMGL/wT6tS22EFwyPBnzyM5/lzt07bJ8/j9OKPMu4d+cOCxtr7Ozusn/rHtPjAV4KDqdjjLGc7B1iphkySTDOUmCotVq0et1qzeWDvZc/HT/yEIBOUj73+Z/l5o37vPHmO+w9uM9k0CeWsLTc5mTvdTR3SVRJ0l3jyU99ms2nnuGrL77Ip//ZLu5uyeCzV5hsd4iIKLxkKI65rQYUJ3s8L7ao1SOQhshoCpXgRYF1hnqjRpxECGbawoJHS8rfLz4Kx+/xXnJgNf1hzvKR5UqyzO6dV7mCJlUF0/wYW464+d47JPECrSji3q33aLe7fPFzn+Wyq7Pz7dfZfvppVH2Vt3bP8d7xGPdgwLs3c2Sjw/JWyt/7r/8W43LE/+Sv/JUzBaZZwfuHmZQ/ON0UgirQ/ckec02/OYBcaYgDxkx57eU/4M2Xv04iMyINSRyjpEcimGYTSmuCU3SUIEUAAJutJkVZkOcZcTMJ+spiBoBJfMXCuHjxMn/zb/5HdFpdlNBcuvIC5y4+ycGDW9x87w32+4cktQ0+8eIXub+7w803v8HiwgK9cxfIp4fc+PZvc7d3iec+8UXWNp9geWkBEZlqD5V8+w9+B5zkU5/+ClZESBEjpUFpQb3RZjis0esssn+wA2jGk3Ew2ngMuPRo+66wjqwym8ymJxzs3efN9+6zt3vAlfPr3Lt3n1q3Tf/wPh975kn2j6dBmzPyCG2IpcSICKFTzl/+GI32BnlhMd6FWEFVZkkuECo8dr6mPq5VexbnzUC6OI4rhqIKl16EjhitQ9fMn/V49Ppaa8myDDh1Mp7PzUc8KfysDVj4eY7mwy9+kI3qPcZYCmORWldKTm6+FoRmqhlbM1ziJIrmwF5WZty/d4+XX3uVP/iDP6AoCp557lkareBxsbq+xngyYW1jg067QxwnQV6hcqc31jIcDKjX6+EeWIPFk5sytJni6S72uHPnLkcnJ2yc2+JjL7xAUZbUa7XQlWAMgpBT1Wq1OYP30VjuUWm6Wc6SJAlJksyv82QScoQfBKSefZ9Z3jNjV84YpI8yUqfTaSBkeE+SxKRp8lAB/+w9nd2jGYDebDaJougnAkg7e71mgPUsR3w473z0NdTrNZRSjEYTTHUPw9wLucBwNODo6JA4jogixdbmeW7fuhNyHu/odDpcuHCJRqtFHEdIH0yz4zgGKRj0B4wf7BElMTqOAgM0isO66z2S09z1cYzuH+bcZyPyAlU6hPOYPOPm917l8PY9Ll6/xuh77+L2jmk+WdBq+sr3IYN2HdKwDnlvccbjI02s/uzXpp+kIZ1i79Ztjo4OaLZWqXuJ8RKlIho6ZSQTRG+J9vkr2Jt3ySdDSlWAckTSB9aktWANYGjIGL+/z5O1Oi7S6DKjt7hAZ3GJfDRlUI5IU8WwyGnnmk4UkyKIC0ERO2zkiPYfcHfvhGEUc97WmSjBJIa1cwsctxXl3oBVkdJaXmLn/g5FVrB29SrjaYbZOaKeW2qlxVtPUTNYY8kO9lFHx8i8RKWh0ykQFANpUVUSfdJKlLH4O3dZPDnGJQrfaNA/2cckktgZhNJI6WlJhTKOAcEvQEYxO8byVjnlAIlQitRJytIwtgYrY3SFTWR+wuhE8D9+6Qs4r4jSBJ0DeY7PDD4KlQuPxLuYhU9+jLXJFDk8pLW6gZVRYFgimNoShJsXqpWShEYETxSdGgwGMqCh0+ly5cpVlNQUeYGUQZe73e5wYg7wzlNIj0okvYUF7ty5z/ikT31tndHBmOFozEK39yPNuw8Fpsbqgy0PZ9mo80U0/GDOSqVqdRZytokpGmmdKNLBpVFrfJSidYwXIJRCaE2c1nDGECUxHolUEmMdvV4XQaXhaQ2OWmg+siXaaNzhIQc7B9QENC9fxHdXmGYGg0AbQWQzrj9xgfXtLawVaGPxBNF14yyRkly7fJnD4xHWeu7fvUe3lZAoSYlm/cqTnBzu8s4br+LLHFsFI05oZBThhEdjkDpoTghRoejWgI+QSJwQREoHdqMWeGcRxuOswQhB3Fpm68nnqHUW0XFCga1kAzSZ8pBPUBpsHFOkCZFWdBs15Po6e/dKbo88z13aoOdzGrVjlPAcHR2TZTlePKxX9KgW7lzE/3Tfq0Bv8Gc2QXGmi2Ou6eNDS/AcYJXyh0rT/iSGVGpe5c3zAocgjmKEVJTGcnh0FFJeKajVakgh6ff7LC0tVS1FFq0UWZZxfHISWK2VlpGUIaDVSs0r3zoKweZgMGB1dZVGo8HJyXF1fR1ag3UGjyZNExq6GZ4PQiIVtGbdHHCdmSHUarU5cKajCCEDQ9t5Hxin3hMnKbN2GlOGtqY4TuY3zhrDNJswnQzxlSv6ua1Nvvj5n+V3fvu3efedd3ni6Wc42NtnbWUVrTRl1WY/A4FPfZ6DfqlzAchTSoJ3HOw/oCgyhsMThpt91lbX6bQ7NOp1kiip2LYCgcIJAd4hRTrXl5rpaQb9smBuFIJUiauAZSECiKuVwNmS8WjC8cEB93bu0S9zXvj488EBdjIhLwq00ggRij9RrCnGJaPxkDiOMKacJ09pmqJUALv+LMzSPsx4VIpj9qw557De40uHkSCdJ6m1mOQlZZmjKtYzCJASa8oqaLcVaGqDE28FoIY1PDDupAPjLAiB8xLrBeMsJ0o1K+cv0Gj1cHisdw8VTFy1ubXbrbkW3dnxaAA9D0B9AP3xHqzFuxCA1mspJDFFOaHRarCzc5OelLSaTZa21rl58yaJikjbTdprq6ysCB7cv8/RvV0Wzq3TubhJrdPilVdf5Z133mE0HnH5ymU63RZRkiC15u6du0ynBZevXqO3uEjhLIuLixRFgaolmNLQaLVImnUe3Nvl3dfeYHjSZ/HcJu89uI8tSkaHR3Tr9aCdFMUkzRbdtVU6i0vBvVfMWG5/unPlxz3+tFv/vx8j6OwzIaWi2Wzw3PPXOH9pjUH/BVrNJoXJyCd9bnz7b/HKH77MQtdxchJx4eoT1DdWeSJuIl6+ge82Ga3XiYWkZTRGOOJanQLF/vCQV957heeefg6h6zgpgymEV4Fh6gOQ430wtwqxUACsZod+lg316PXxCEoJh0y5mwrGTUVzaZNy933Kvds4e4/Prmue/+U293Yd770tEKLLcXHAg933+Tv/zR2UBJFGrGyfo6Pb3Nx7m2Rg0UWEbzqWWg3qpkB3l3nq2gu88PEXmYGDUOnSfh/s8wcxsh4/RMW8/cFsrp+k4YVBeBXYow7G4xP+8A9+k9s3v0esSpJIECUq/J4IJjRZkaG0w7mgSem8YXNzizybsLc7YWtrmyhKCDasKvyr1kAPJFGNZLHSrfQFCI8WkvWty6ysnufWrdt4f8Kbd+5Sr7dZ27rO6soCyJT22mWaK68wzgqWts6xurmIFCB8EnTzveR3/uXvcffuHT756c/OC7RC1dGRx9uYleVz3HjrO6RJg1q9zauvfoePPf/peTfH9wNTvfeUpsCZkmw8ZTjM2T8ZMJqOGWYTfu+7rxEpz7mtZZ57ZpO3X3uF5575FMdZzvH4kHoUWgtT1aKzfpmNc09Tlgrr8rnDuTE5Ye6mVTt7UTGsH2ZtSikrXfTQzm2tnYNSp/GGYjgcoHQgFMwK0x+VMQPcyrJk5j0wO+6zLeWPrrn+ka9P74+by2+E4mdgf1Llas46pD8DwspgouQJbZaz+z+LP775zW/y9nvvVl1SXVaWV3jqmWcQwHG/j45DThfFEXESU5QF+7u7NJtNhuMRxoR4oV6vU5YleVlSGoOx4VmKk4Stc1v0T/qcP39+DoafZZbOzEZP28IfBvTggyD77OdnYz5rLcfHx6RpOgeRz7bXP7qGz+bQWeB29n4zmYEsyxiPx8xa2IPWvwyxRfXeZ+/hWdmM4XBIs9mk1WrN5/FP6nj42EOB3525R955GvXGvPAxHo9pNps4YymLAo8jSWIGgyFpmtJutxFCsbm5xeHBIe++9zYbSUI2maKkBFdDxwm6XiOqpaGbrZZymo2GuMGUBiUihJqZMPIBIPVHue5xLeXilcu0O22SKMLkOcXtPUbv3Wc0GWMUDN6/z/jWLrv9I9rba2w99yS+Fgd91er5k1pWGuw/HR92SKG5/dYbGGfIigIxzZFpC+sEzTTluFYjFyDWV5BJxPjkELO3Q3M0JREejUM5R6kgVxaVl/SMoadjTC2mwLPQSImTmPG4YBJFZJvL7NzdYcF4lp2nIRU1FaOcoVZCYhxaGjSWYX9I7i2mmbIYtYn6QxJvGXcSOlLRHxbU20vUV89xdHDIsONp1WJOjo+oZyPqfkpUSBiX6FGOLkrAYIXGidC5Kb1EGhe6EolpHk+58/Vvs5Tl7OqC/emYkXOU0lNXGj0uSZMIFwf2aCoCZjJEsJtPuSstgpJERihnKT3INApztSzRFXlrZW2T1dUNhtZTCodLNbFzkJc4DGUCUgkoBPtpROcrL1GbTijTOoXUgbhX4SEI8Ge6DJTUWOkIxtinwetsv+92e3O2vzGWojDEcUytXmc0nFBogZIR288+zYNbd/BRhI80pXAMJmNaovcj6Z5+qL/VZzaW7wc8fGBTP7uJeY/SmqWlJVaWV2m3miRpjFeaHM3u8ZBJnqOiGOMdtWaDYhoqsrISpM7znEazhVAKpcLF81LiUEzzjIY1FP0+qt7EL3cZ4hiXnnpUR5sJg937fOzTL9JdW2bsCuol1JOUUoC3JUKFxVxYz5Xt87zx9rtk+YSb79zk6rWrGC9o9Np87Ge/QFbk3H/rNVxpggNjkgACUxgi7bGYwEKsEmnngqC2iCKUELSaLU5OjjEl+DxosemoxuLGBZavXiddXAGhMC4DHNLXcF4wtQ7KglouKYTCCENTChLhadUTzPo2D3Z2eP8o54Xzy9TTFKVjHIqDw0OKbDIPBMIEfRjgoKrWn+1L9RUTTswsA88ArbP7PtM/eqgKWRlv/TiH954sywjdPz4YMSUJURRhTdBfkjIYFQ0qkLQoCqw11Gt1sjyj2WxQmJyj4xOiKGY6ntBo1oFw3Rr1Os5BFKUYFyomw+GYoihYXFxmNBpTliVxHK7xyck+K0tLxGmK9Q55RpgjPB+SuTSAdwipAqPNOhqNFtaWJEk0T9xLY6sWaolAYkpLv9+fV/lDa1hGkU9AeJqNOkWREekuzUaDJInZ3z/ijbdu0Gh3ufrkE3jvSZM6JpuE9uT5BQ3/nHAB5KqOeVYo8N5zcnzEZDLm5PiYk+NDlpdXWej26Ha6NOtN0jhQ7qUK1VqBQopgVDbnQXhXFYWrgFV6lAxaKKFCaxlNcybjMTu7u9y7d4+bt95neXuL8TQjy8bEscY7S7vVnjOupQStA9O3KIp54D1bcOVDDOGP9jir/TWT5LCVprSzHu0I4GpWBJ1UIcBLjCuRCKwocFUxhRm71Yf2flOxAKwD5cFaiZcS46DVW6C1WqfR7eGRFK5Kys7okM2e+U6nTRxHAVRSEme+Pzv9oVZ/51AEd9vde/dZ6S1gioLSFNSaKdMs5/LVK/T3DymtgUhx9fJljvePaDuB0QIXSVZXV7n79rvcP36LhYtbdJY7fOKlTzEaDNg/PmJn9z69tSX6/T6NZovrTz2NtPCdl1+hs7DIrZvvs33xAq+++TpPXb+OUxpdT0lqNd55/U3KSUbcqnNv7wFRFLFz6y6RkCRaY8qCKK6xsLFOe22NpN1CaoUz7qfB8A89PiSN14WCj1CWzkJMp7OMJALVRrFOJ/oq3/vur6HTgtwaFs5tcftoTFKk+ExhYomzAhHVKAAlc5QokKJJOdjju1/7x5xfbbLQvQ5CMm0Ykiwcn5ChLTsYZlRFSikCm1HMWKrff1gZTKj84TFHomB8qcf+vV2GqsU7h4by6DW2z/c4t7nEyuY5Wp2UyT7s75XYw4KpSBBliS4c8vU9JuoujbQAF7pc8sjRXV9lw9T5yv/o32N4fpXlpSVm3SanRdUPXus/Phga1pw/isH6kzKEr66WKzk5OOBb3/iX7D94lyQqSSKQSCIdgfI4G+57FCmklmhiHAFIGY4GrC1vsLa6TpqmID1Ce2QkKuklBV6iRFTdjiAZEdTGKg1sJygKz//tP/1/8lf/+l/iky99CjvJuHv3Fnfv7aK8otFco3fhGdYTxfLaeYTQWGFwXqOcQviI//n/7G8wMSc4aSsTwwgooNLAlDiWFtY4Ojrgwd27NFotnMsqmafHFMM4ZTwaU2LKYCLrfTCijpIYJyQTA4lQ3N/bR7wx5HPPfpqdW++zcnGLzKc4JIiIqNbl8hOfxJHiXYkpDVGkg5aa14QiwKyF+xRIne3hoY0/xhgzB1LjOJ7vN0qFlGc6DT9L0hRrS6Loz579dfZ6eh+0XAMDxz50no8DU+fAKWdC9CoZDYZp8w+pgOhgiDpPVM8wpb0P81lJSVR1Jfb7fbrdbjDyrNf47Oc+x/rmRiAeHB1z/tw5VhaXAFhdXub4+Jh2q4XSwU9hJiV2eHJctbnHDAbBF2M6nTItcpqtJv1BPj+/OI7ZWF8HHzwpTFnOwcg4jkMH3BmJprPF5g8LrIoqFr1w4QKm6mgbDodEUUStVvsAC/Us63QGnAagPsRT/X5/Pu/SNJ236gcJAvMQKDszYQKqvKTkpCJxtFotWq3WBwpyH8Xx6LWeH6//IJP9bKw9e12v14njmCgqGI1GLC4uMR5NQqdTpKjXa0wmU/r9wbx9V2nF6toKjU6TuNJDlhOJUBojBJGIsNOcSGpaUTQ3ufV4irKsyEEOkH8cMupD49Fzd84xPDqh3z9GxhpPysr5TSYO0rGjd/kiFrAeJtmUxXMbrDxxEVFP8FqBDIQf5RWR1JVrxU/Hhx3OC/Zv3iQVkqW1dRqNlL71lN6TpBFJGjPOpzQW27hEoxdq1GsJvHeH4aRPHAlSH3Io6VxgefoSTAFOEwnLdq1O5ixZI6KdLBOnTZRKUKvLiHqL+zdvU/PQM5pkJHCRxgDKC466jvrEkfaHnLg+tcM6q40W/V6daDxhOUnYjzwqSdjfP6JuYeHCNntNjR2liP4R07FB5FAzAcx0sw7jCr+ROKT3WOGx0pOcTMhu36cjJZGQ7OzvYUxgeyZxRJOIVqPLXnFES4TSw5H1HExyphZ8FNPwOaV3ZOMpSsc0GglZUaI81Jwi0ik//yv/RjCRx+HjCFKJNIpYSkalJXc2sFu9ZGhKdJoQJQm5I0h3EIqqivB8+GrNDRiTq9bfYDitlA57eJKgtarkHS1Shj2nLEuUhHq9zmSahTjFK7rntnni85+jubaCVQLSiP1Rnw2x/SPNux+qzR8eBk0/8PUjm9eZH7LQ63Ht6lW2zm/TbjdJkwR0zCh33P5Xv8PNW3dpd3tsbqxz7+5dkighLwsiGXyoa2mKsWYO7BZFjvclOtHURRMtHC6NESLFDAwbl1ep1VP69+5z/82X+dKXvkDcqlNMS7ppQtSIGTmPdgpBEKnGCRIRsdRpcXFrlbffu4nUCSf9CZ2FDoUt0M0Wn/7Cl/iWc9x+6zWsyQOd2lu01BjjyIucOEkCSERgTBZZHgyMZBC/3tvdxQEqTmgsr7F+/grd1XOYeoPCOFReIHQFfIiSXAhkHiFlQt8XWBSps8RSgBToWNJq1skXlnjn3j2k9zx7YY3echSa1YTkYP9BcE+vgg0p1UP361H22GxT9ATQ5fTnD1dp4dF0jKqV8Me7CRweH5EmNeI4od1uUx4fc3hwxNLSMgBrq+tMJmNipVBagRD0ej2m04xOp41SEmMNeRGSi9IYpA4s6vCgGuK4ESQmYoFSonqoJft7B2xubrC4uMSDBzvVFfG0Wg2kOsvo8Mw5nzNGb7XJC8lccy+uNJWSpIasRLKdczgbNFGIQhvieDyea2oVRZAcSNOYVqtZubEalJaBWVsapI64/sxz5CaYXI0GAzbXN1hcWGCym2O9mx16dbTgvGHWMj5jIHlEMMRCMJlm3L1/l5OTIxa6C6ytrrGyvMxir0e33aZRrxFFutIFggBMM68wVTE8kiBdIbzDuRxTlOTFlNGoT38wYP/okJu37/D2u++x+2CP3oNdukuLdDtNOu0mUgis83hvkDIYdSGg2WwxHodqf1mGQD5IIkQcHh7T6/1o9P4/7XGWmVpWCUWaBvDcW4eXktJa0nqD4/EggExi1n6m8bMAVszkOWbvKSqWisY4j3GCtF5HKs3CwiJxrQEyyIxYFwzepAiaU8wY6N7jSkMtTUjiOABKs6D5Mcy3xzMaochzBILVpSXyScYrL7/MF7/8RQaTASpSDKcTsjzn5E6fxbVVqDdYWl9l5+ZtFno94m4L30zZuLzNK9/6Lrfv3uFnf/4reOEopKex0OZqr0WjllJOSprNNr/x6/+Mr/78V2k0mkyyKe/fvsXWhbA/WROqmtuXLvKd732Pa1vn+dqb75IsLzAZDujfP8aUhloUYfOctF5jYXmF5a1zNJcXKQTEzqHlTIP4x78e/us8fNWGJLwmcEI9Fbkdh8X4iOXN6yysPwmv/S5aC0bTY0aHI8rRFIqCpR3H9nvHvN9pYGSMUxHWW0ZRjG60KG3EP/tH/4oLxVu8sH6O8XObDFs1dBSR1FKSJA0FpirwdtbiRGDyCz4YLz08HFrC5eYCxe636D1znVd/47dYO7/IDhu0RznlyUV++x++x+HwHtTqtGzMxfoiP3v9kzxx/mm2Wi3qhaEmlxDLDV6/9S5v79zm/skO3955ndfe2OGmS/j63/y/8vSv/Dn++v/wV5nLNcEpI2c+zr764wOif3KeyD/+8VDbOjH4kp17N/j6b/46xfSEWiwrGRrFeDwELUnSCOEFsdTIJArrpQdEMD5qt1usLG7Q7SyilEbHujJWCAD87ffvoqOUzY0tkBXw5SUQEdhcFikMcWL4d/7dX6Hb7kFpAcf21hJlXsdQI9URqtnGRx0wAaCVIsdhQTVxXlJfTEjFSmARU+n7Cg+SYHqpoNXscrC/z8nJMY1mnVE2oBb15q3Jj2OnGmPwxuCsoLAWUw6QTMkd5C5GigJrpqi4SaIX+e6rr/Hx56+zv3ef3vIqg8kIk7S4/tRLINqURYkxBVJJrLGh/U9owIIoQ9IoVHDr9adamN4Hf4U8zysz1GT+8xnDczqdzp/L8XjClSsXP5Jr8wx4MsbMY+2QLJ7qcc6BrBm7Ucxr4CB4CLCbaeLP3ifL82AeWsWnMzMqqaoigfcMBgMe7O7y7rvvcm57m+dfeB4dx1w4f57NtfWKqGBCnuNDnrG6tMJSb5HeQi98lguFhqReIyG06Rd5zvHxMXmeM+j3uX33Disrq2ye26ritsBadFi8dWSVSViapvNrY609bc/mFMyaJd+PY6TOfm/2vdnvn5VRCGSI+ANs57N/fxa0dc4yHA6YTAJZpVarEcfNQPipno3Z/FNVZ1tRFA8xXKfTKfv7+0Effn2djY2Nj2zH1NlxFjx99Ouz+8zMH2H2epb3OOeCkW6nQ5aF/MWYEussZVlS13WUUrRaLYwx9Pt9rLW0WnXSWkpRRjTSGs552t0OJR7jLDoDOZ0yvHdAL2mAVlglyCNPutCBegzK4mVVROD0GfnjXoPZ16PRiGF/SOQF5SSjKC0Yy7SmGI4n1EtDFMVYoPvkedrry4hmDScFCoGedY94P+sh+ZHv03+fhlWSwe490sLS6LSJGwl1XyMThigRNJopg/0DVK1O1FBYKYm2UqYezLDJ4d4OHaFoIEl8iRCOkS/xCrwzxBNLWwg6OibWik6tgzcRHReTdluMpWJvMuby+gbHoyHTSU4zjWjGmlgJGkoRuwkyishkicim+MMJquwgSsNqFLN+bp08z9isaRoqoZXU+K137/CJS1dobV1htP+AiWzQiiKMD9q8IJFOBHk74RFKkBuPLafUI0/WiIi0IJUpNs8QIqYuUnId07mwwsbVq4zf/C7l4QESxQTJcW6QMiI1DqsFo9Ki4lbopBlnxBik9cSqTqu5yMbGBsVoSE1pMlNiaykIR5ymKO2COa+BTBUkUuELi1ehWCycxxmLEz5gFi4A4zOi1axbdbauBrkeUcn32HlurJSvgFZHllvqaUK9XqMYToIplo5Zf+FZyghKBPV6jQf7e0j7oz1nHwpMjaIzejCPqUL56uuHE4hT0C1NU86f3+bJJ59gaWONei1F6wghI/aO+uw+eMDJcAIyYmmxoNVpU2QFsQyB0NHxMYuLS3h8xc7yeGuxCozLiKQCmSKFCM53cUyJZP/mO9z/7rf51EufZO3CBepRgvKCzHuEk0TCY8ocr6LAgDUOayyxlmxvrHF4dEx/OGaaG1qdBgoHWqPSOi9+/suY0nDv3dcQJqcopngR40WM0kEDs/A2uDgiKI2jKAsSrfHGkKYpUb3G0oWrNDbP43SNkQfyKTbL0SLCigjhHFYZ8KCNp8xVEOIvCxItKAuD1RFalMRasNhtoMQGb9y+y3Sa8emnLtDpLeKdRyvP/v4ek/GkYqEy+0/FdDxlpZ6Vbjgj6BB+NvvfrBL+CCs1aCDO3vvHN2agfxzHgZWapoyGYwaDwbxC3Gm30TI4OOZlQb1Rx+Mq06eEyXSKMbZyhC+xzpGmNUajId6LALKWBWM3Jm3UMaYkSVLGowmDwZBet8d4PGI6nSBlqKBz5ko+1IY1Q7YAJWQlT1CSxHIe4M1c2ENlBqaTjEa9gXeWrCiRAtI0GCwMhiOajUZgqeIqiQGCUysCrWOEiOgtjvjcz36Jgwd32ds/ZGPrPC99+iXiV17m5u33KEwoDsxAT1sBY0IIpJeh8iVnAemMOeMYjyeMR2OOjg7Zud9hZXmZ5aUFep02zUadRq1OGifzVrWZwDziVMPPe4cxOXk2Jc8mDEcDDo8O2Ds84Pbde9zd2WOSFUilefDggO+9/Aq/9Mu/QDad0Go0UDJopCVJhLGGRq2Bl5KJEEynU1rtJvW0xng8ChIQQpDl2Y9xlv5w41HwccaynYyn1Gs1amkN40MCLoREJillWZBNJxXTKJjgaRS+EvD2LiRfcZxgrUdFMSsLi+goRtZrSKWxzmNcxSgJGQuiohsJ58iLkqPDQ1qNBu1Om3o9sDicr0yuvg+L6dFz8T60dRRZRqI0rU4Hmm063Q7v3XyfpeUFhJDk2ZjmYhd5cMzk8BgZR9TrDbYvbHP7rXdZimNMMyZuN9h44SkuWHCFZe/4EJtlNHREq91iNBxzcNinnjTotbu88/7NwG6xjnqriZSCTqPJJJ/S1jFvvfsOi4uL3HnvFWQc8+DkmLLIGR6dkNRqeGuRUUxcS1laW2P1/BZRo06j1araVPyPexn878UQMzAIQEiEV3gf0FQjHNNsyn/0f/ibTE/uEJmYlki4f/8e72djymKKScANRqx96zYLWwvsL0mQMRmCaeHwzTq34h4v1FfYWNykfO8eg4NDjq8tcjjq011a5I3X32Q8HPBX/p2/zNLCQtC9FGdYYNV4lCEFoLzEWc/25WUufCPm3t4e209doDCgVp9mvaxz7cov8Hf++f8d2Wtx+co1Xv/aH1JXnmz/Pu+cHHFle52mNfT0IlflJT62do4/d+kp8prin775Db5z9wb/8u03iDZ7/PY//sd85dp1nl7ZCIVsfLWrBwYRCLyoJDd+CCD1cSDFR334+fmdxjti/t3Z+Qu8Mdy6eYPf+/qvgTkhjQVxpNFSE+sY2fJEsWJu9CcEWnikmiUBgkhF1NMmceJBGCJdQ1ADIvAageLO3bvgLWsbKyihq73XVcc007UG4QQvfuxFnMkoTY4TlsxoJG28nTBSgttvvMHq8ioLyyvIMhQnlfD4KPy+lwbpHVIkVQlCV8CtDTJQxMCURr2BKQ1QIKxBJg/rdcLDIIlzDoPDugKsxTpNbsCWY5wZhWKHiol1TLO+QL1WcHvnfZZ6K+SjATrRrJy7SmvpAmWhcC5npkfsnEAKHUIlyZxpKaoq7AwwlFIyHo/nzMXZsUoZZJpmxcgA+ARW5sJih16v95GYu4/uk6Yq6OV58ZAJ1VkwdTbcrKdo9v15U6D/wGcEgJNg9IQLFCB8VQS1lTPyhJOjY+7cvs3x0VEAqbMsAH1bm2ip0EmKxxOlabimFRi5sbpWGWaa+b2x1tKo15lmWdUlFPLCBw8ecP/ePX7v936PL3/lK6yvrRHpKAC7gClKnLVopYlneqM2JM79kxO63W64bj+gu+hxDNWz35td29mxLiwszK/x4eEhURTNv3cWpC3Lkul0ynA4AKBWC8aqcZxUYK9hZh40m29CiMogNfx9kiSMx2OOjo5oNBqsr69z7ty5OSt2Nq8/quMHdjGcJducAf9nsf5sngohg7bhyYDpdMpgMKDd7nBwNMb6BOctaVIjTWLKsiSbTlBSzouZWgcJj/6gT63TBkBZj+mP4GCAFDk4yH3JsSo598LT+MST50WQo5MCJQVBzsw/NE9mx/5HXYOzz26r1SJd6mKnU1qdNp1eh7qOyGyJch6ZWYQFjUQ1apQKKItAJCH4algRnkfdaCDED7Mj/3R4AZEzeGMZ9gd0raFWT/AygHDLS4tMdo7whUVEMTJ2ZJFGXjhHfgfsZEppPP3RhFQ4GkKRSIHXHofB24LR4Jh0krPUXWBlY5G+SijXMmy9RuEcutmgs77O7dsZh5MxiyRkmSUqLT1viWJNrqFWlKhY4JSkmRuygxOiWBOrjCifcqkZYyNNUUxY660Qt1bYlQq9usri6jnodMiFJLWeWkXEskoEKTYlsIiwvrcT6k9sk926DXmGFRZnSogi9IUtnvzFX2a11eXVd15BaoErPa2FJfbyKXl/SmIiToCyluKsIs+mRJKZACAqibh0+SrdhR43Xn+d1VoL16rjum2cjHAyQilNojTSOMYiJ1hmSbwzOC+CLr0Ai8NYT4U6hudyJlOjTjsCvHekaYIQUJqSyXSKFKH4GoyzZbXnG2pJymQwgcKGgkuqcARZOy0lo/EYJX60dfbDganqUVbNmbaJGajCjI47aweeLaCS9fV1rl69xsbmFq12BxEJnIxRXuCKXQ4HJ8SdJSZFxngyodFp4VEYE1pzdRyHFunqsz3hwntTokUN4S0KT57lRPUak2zM+M47jAZDNja32XzyeZSKySvNyxhPPh2j6zWcd9RUFWjZEusdxgcHtCefuMb3XnmVPB9x+9YdLl+8BEUJ9SZZlPDxr/w5rMt58O7rRFJQehMSOicwRaW1GcVYZ/HOY4ynLCRaaRY3L6AW19C1Bvkox8tifr1KaxFpg7LIiSKNjBSiMJSyMuKJBCaPGEuF8RmJd6RSIWSorna6LQq7znt7+6T3+zy10aPdtSjlkVqxu7PDeDTEzLxhhKx6gqqNz8sqkA2fJ72bCwN7qJziKyD9kSDm0VabH+dIohjvHZPxKAjTpyk6UtTqCVGkGE+GZJlieWkpOJOqAIy2Wi2sddTSGqPRhNJ4nJ+y0OswHU2ZTqbkeYGK6uRFqJYbYykyT1T3KCXQkeTo+JB6I2FhoceDBwXWegb9EUuLi3MGBS4kSKfPSmAEqMq06Z233+DJJ58EIIpCy7R3QRR7PBhw+/0HXLt2hSwrUQrSNKYsBXt7R9QaDbqLC3gRNNsK44i1CoxXACmRUcTC0hqj8ZST4z537r7NhUtP8T/4i7/KL/zir/Af/yf/Ma+99TqOgmDw4FE2wQtXMRssYn7vLUIQNILRleGzZ5KVjEY7PNh7QLfbptdpBUC12Qh6qnHQ+o3jaK4T6yuNsKIsGU9GTKYTBqMhh8fH7B8eMRgMGU9yhBAoFSHQKKe5dese93f3OLe+TKteR3qo1et4AWkUIx0oHXSWlY6CXMZ4Siw041GQB1DxR8/J9+x4tP1PeDBlwdFkTBxF1FrdYKagFe2lbZwtiScTtFZzcKTqzEdKRVEWRFFMu9N9SBvMeU/pAWPnIL88k3jMCyZ4Dg52yfOcjc1V0kYalo9Kk3VWUXdUS4o7LcwAp8U3IRDeYccjGnGMwzPOM4o8R0bB+Xp0OODC+QsYUZDEKbWtDcp8irAl08ExaVpn+fwG77//PlvntlFJSqfexRjHST7BWUen0QAdtF+zwvHd77zCSm+dRqvNZDKms9CjtJaL2xeYTrMKCPWMRn1OBsc899RzvGJK+jZUUe/vPqCRaCJTIrXEx4rFrU0Wz22gOi3aa8sgmOt0+ZmPzZ99vv6nOj5se/ifzL4Q5EassihfEjmFQyBEjhWCW7fv8l/8F/8fPvWJHlE8ZOL6vPb6AzZevMCb2R0+/hc+SXbjHlPvaIxHnCzWKLwjs46yKDne2WPYd+zYA/6Pd75NdxCzdgKL7yzy7M99lvzYsL6wwRvZiK9942v84hd/HtVsIEUQIX7UyO/R6yO8Q0gLOuLfeulz/J///n9NK044d9fytK2zcm2buKH4K3/9LzNuJVy/+BT/4R98m3fzu7hDiEvPt/bfRSWKjcU68eg19HSBF3tryOmIW9N9Xr/5NrrT4m/87/+3vPXtt8hVizv7x1zZXENX5j1WqEoX1GOFQ3nLTFP1jxqPK/T8ZIxHj3NW5HQEyZ0AZL3x8jd47bvfQtqcWMcksURHEikdUhsSZAVIOpyQOAT1ZgtjHUppJpMpSappthrUam0iXcMLidRBUx6pcFLx2c9/ORTdvQPsnFYYIgWBxyB9aPf2pcM5g3AO4QzSGpwP2v2+KPmtX/8npEnKX/1rf43CF3gkUkRENqUoHPd2djm/vY1MfBB0thIhXNV9EiBc6wKDbmV5ASUEtixO3dN5uCPJV9rZs9ZwY8BagSkLppNBMH0sLTiJ0J60rpBNS6OeIjKBNVMmjFhd+BTXzv8c1mgKP8Fah66KtAHvN2c2kEpHsgJxz4K8uuokOss6VCoYZgghyLIpURRRlAbnDefPn39oD/wojLPs0bI0tFot+v3+nBn8KAvQA04yBzPhlEnpKrbqjAgjQgPmnNlpnUQLCV7jSsN0MmEwGLC7u8toPEZIydLaKlpraknCzs4OiwsL1Gq1h8y9Go1GILIASRIzmUwZj8b4ug8dK4Azdi6z5PE0m02yLGP7/HmWl5ZZWFzAe4+qOAbGWkxZ0mw2523wzjmMdxhnGYxHNFottJyx+EIxF/EwKD3LRx53nR8Fwh7ttux0WkRVXtHvD2i1WngPo9GIwWAw9ypI03QO4BsTJNTCWwUJq6LKB4HKNCXo+B4cHMwNbDc3N9ne3p4DqY8ez0dtPE4y4VF2sBczGbMg/xLmoa20HX3llyZoNIJ5WRonHB0c0ul1QAqcA1uC0J56I8JFnmlWkGc5xdSTGUsdj5lM6bRaSAdeK5wSTCJP7+IKwkcY64giQWQLpg2Nk57cGOK8INYKlEQp5sSVH+a6P1oEKYqCo91dhsMhevc+jUaDtbU16vU67XabqFOriC0BM4nEw/IRQgh8WWLLMszlP6kb9t+Tob0h8hYTCfKdE+TOMfZCG6XrSN1FNSzLnRa7D/bxkaCWxowmJTRaxGubxEmL/PCQsTxEu4TRYEquBcp7moWjLgROeQpTUB4d4F3OwqWLbD39BO8Oh+yPJzS6XfJEcWQyfCrZ2F7n7q07NMYlVkq8zWkLy4KVTJMgVRY5h8ThygyTT/DlAaom0N0OiYYXz21Sdtq8c3xAY7lOstQlaTSJXIxyMVo6yqrzVXkV5CNF0N21ImHlxRd597s3mL77HkfSUC43WN+8zIu/+m/T+MTHOPrma6jhGC1KUl9n84kXeO7TT/Kbf+fvkdw+ZKmckuEZFjlTLfEqdCW4yOEoufbUVaw0mAf7jM0h+6li61OfYuKhtCU6ErhY4ZBoY8mlg0ijhMCboirvh8qB8AKhJFqHGMCUQY5GiNPibb2ehiKpLciLLMTc1qGEwAgLUSgml7lDajDCgwuFap87RKRBKaxzlKVlkk3ppH987fQPp5laiWCfXTQ+EDhXm3VIISsWk5LUag0uX77E1tYWnU4HFUVYPHiH8zCaTHE2OI95FXEyHLK8usKoPyKKEpwzc1fDqHLgcy60uigT4dGUyuOVZVoOiK1HFJa9W7epLy4RtUMLsHeuahGZgQOVORSS6TQLjIMqCIuiCIQkihOuXL7EK6+/Qf/4hL3mHmtrqyTO44Unarf42a/+Bb7+G5rbb7xOJAEMwgaw0QBelgH8dVUHl/A4KSlNhh0NcHmG0Bqh9JzJKXREbi1OSLyNcC4EUVEUUxYFZV5Q1mtYV4ntWoenhvCSNPJESrLc69JIIu7evUcvFpxb6tKIEkTcRMqY3fu3GRwfBhaELfG+DC01osJNCQB5CL7kPAjDg3QC4T9Y4Z2NH3Yz+pMa9XqjCjoLFhYW5qL5URRhjCFJEsqipN8foJSk1+syGg3nc7soS6bTCdNpFioupqDdbFdt/5BlU4yRaC3AhWq7co4oUiRJymAw4PDwiJWVZYbDIaPRhHq9foaxWzF/zhQanDMhcpQhALl+/TpRHOFFqJYWhQltgU5wdHTEW2+9webmKoiYZquOkBqpglxBs9MBQiqc5zlagBQSOWNvuBAsBNH/wEi8cvUa1564htKaOIn5+Mdf4I0brwfjokqHRSmJ0ApvHdacYS4JGcSkRQiKAnBWJVwqFAUODo84PDhARwqtFfUq6Iy0JtKBoWqNwVZAal4UTKZjptMpWV5QVomaF0HbChG0UqRUaBUxnkx49bXXOLf25fB8lDm1RoOiLBFSkE8zIhEhpWRpaYnJZEJaq1FMp6FzUIiKgfPRHWer+lLKIKqvFGVZMh6P6Q+HSAGdbpeFbmDZhPuZVC6IsgrKQhXeV5o4AWB1c5Vc7wkbm3Vzxvn8B4Rc9uTkhDzP6Ha7LCwsVJpop6aDp63D4qG/rU4ECPPTO4twDpPnnOwfksQxaaNOaQ1Jo44SgqjbwQO7e7tY5anZGqN+H10LGnjTwRCNRCYJ61vrHO/tkRc5xkM+LWiv9EiTmFavy7jIMC6jsJZPfOITxGlCs9lERYqXX351rsm2vLIE0jOajBgMR1y6cpWsNLz1zts4Y9i5u4MyIRE1SlOLUs5tb7O5fY724gJRLaHRbgXztYoB/FNqwZ/OcGLGxwotzYigq9hwCXfeukE+njDc9Xz1i2vkowH39+7RLTb5u7/2n3P47JfYeHqbjXSRleGUYVGyqxNyaZi4ghvvvsflpuXjLz3J9T//eZq2Tf7eXe7fv8vbhWGlvszvvP420eY2auU8Pk4QwiMqSYc/Coz0QoCTLNbb3ItLnqq36Gxc5l+8/C+4fu4849u32N+/z3vrz9PbfhqVWX71M5dxtSFlbYUbbw14935BYTx390eow4y6G/H3du+RMcIVOT72JHFOfJLz6Rc/w0g73pRT2sKwjiRX1fruHdJDbD2lfOjp/f7H/xMDnD5mVEXjinLPaRdVKHDm+ZiXX/0ON773TTAFkbQkSUQUC2S1TwdQIKx9xtt5fJRlU7KsoNPt0u22abfbtNst0qozRkVhbxcyaOPNWoRmJITw+pQ2QGUKSQWuha4CWS0rCmdD94xzHq1j/tpf/asoFSGFRBHYnKF1OefGjRv87b/9X/GX/tK/zbPPPF1JZhUgLIoc7wzChdeltwynE1Y3VyqJlzBXTiWimL+eAZreg7Ulzudk2YSjwwHZxGOKIL0Ta0272SSVHiETap0Wo+F9WosNhGiAblG4Ea4sSXSDsswfArvOxpWPA8dmxpYQ9szZ18bYig0YnIGddygluXbtSdI0+cD7/FmNs8/UTDd0MBjQaDTpdDqBlZdllS/AKRvXe4+VM8YfD7Hr5rGDFLgzT7WbdZZ5T2ENtWZCa6FNsr4+Z09OKxOlwhqUVJR5znQy4fj4uMoJA7v08PCAPM9pNpvoKGI4HFbdQQnDwQDZ7VYmU8UcxFUqFPmXl5c5OTmhWasTJTFpkgQjKhMYet1ut9IbLeZzDREAyc3NzVNA3QU/gSiOHporj2p0zsajANjZ6zW79iF3OPUiUBX54uTkZK6VWq/X0Dr4AsyejVPmajFv4z9r3mWtZTQacXR0RBzHbG5ucuHCBRYXF890sv1kjUdB6LPSErPi+alXx+mYLcX1ep16vR6YvoMB+ThD5Y44Aj8tmBYWYRxKK+ppM+iKigLncrwrmeYZ3cVFoloNCSRI0u4CjbTGZDqlzDJULaEZa0hjxpMxpSmDF0QcQRQhhETKh5+dH2afm93fwWBAr9dja2uLOI5QSpIkydzgrCwN1vq5TvOsozI8p2FhDQzn+LGSFT8dP3hYqdBW0BCa8qTPya3bLKytYaKYMokQi23q26uY8THT8ZhuXKPQnkIUdDot4rTG7XxMLBYoNFh1QlwW6PGYhtOUymFxaGdweMx0QHbjDer1HS488zRr7RbvTTLismS92WFc5qhum3uvDTjXWYTFRSYPdvGTCVNKYgRpPcZiqQkQosTJYKTsJyVmekTSjUlrDqU8T/YSRDulFku8cgFEtTllEuOhMp0MpBeNxuUeKxTNS0/R+9RL7A8nPPf5F1l54WNsLpwnWVzigYcROWM/oekMqtZg6cJltr/4BZYubXPjv/o1Hnztm+xnQ4wneFfUNUJY2g5UI8I9uckAz9F0yta5S+zmQ6aJonACkZfUdYpxFisCpuhsKK4JpZEy7NVn1wghxHzth7B+B8lFOy9eGWMwucHmBm8cBhA6SAXKqjO39B5fUq3FoTPTuQCsChRKBU32wXBAp7v4x553HwpMfTSZPwuinf0dNasQilARlUqzvb3NhYsXWVxeQsVRRdcFby1eCu7t7YP15KMBtXaPk+GI8WhMFMeMx+OwWYmghTju9+l0ekjCzUjSGI9HCYEEGp0ORktqcUKthKwsuXjpArU0RUiBMUFovigKkjiaa+5EcVKZ+gTds5meTWkcC70eW+vr3Hz/Fg92d2m0myy02wgcUwNTEfPxL/wCnpS7N14GWyArZoexQQBYIVEOFBInBDKGLJsgS4tOUiwCHSVBW1MIojjB6hgVJXifBhBTSkrvkVYHsMM5TGmYTjNazRbWCHwtRtZi0liSakncahGxxvG0IJk6dAlp0mFhNUJrzQMpOTzYq8Daqp0MjxOnYZer6NKS4OAuqgrwnCRwJtj9swZUp9OMOE7Q2s2r+L1eD601k8mEWq1GEic468nzKVKqoF0z2+CtpShy0jRhOg4C6GVpqNVCW97hYR8p4+o+KawPjpORitAyMGEHgwHdbod2p8NwNMH6ypyLhxl51VVC+EBRVypFSElar4UKS0UUljIkUThLs9kkiYPDfZKmCKnxhDnTbHfQSiGEJc+D3miaVMxPG5ItKRTeudDOrTLWNzZ5/oXnqaVBn1VpwcXtc9SSmNG0DEzOM/qY1s1c9s4ETkoGIwsIgS0eHEgdh8AE8NZQ2hDw9PsjbAUg+2o+uZDxYf1pUiaqz/FVIiCFRkhV6aTI8LWSOAHvvXuTu0/eo9dqYp0lzzIazUZoD6ullY5nxYowhlwWlFUwbIWjWbUGfRTH2WdLqaDBZ62lLMJa5p2DMiPPc+7dPmHnzh2iKGZ1bQ0hFUtLy0HEW0mUDuzh8LzPAsYZbTmsAMzB0PB1keeMhiPKsmRvb4+FhQXWN9ZotBoIWRUEvPvA8TL/hIqsGpDW8Loyzhr3+2STCYNBn9FgyNUrV7BZgdTBXKLZaHBspgzKjHdeeYsvfuHzpLUGhc2JtKbT7dLv96mZBkkcs7i6yEn/mKOjE7LpkHxnwtVrVxmPx8gkxhSGxWaHW4d3yYuC1994gxc/+UmazQbNRjBBQHimeR7mulSsrG/y9//bv48Djg+PELlBlRYJqERx/tJFts5v01rs4SLF8uYGOo6qFVLMzx/xrz0x9U99PLSfiLCiChzOh0KkVQ5HTOI0e3u7uFhRWEsnHvPlZ1N+7Tuv8dbXS77wpV+isJ73773Kgavz/BOfYnA4QUvoCwfDlOPvvMKTn64zkSU90eA//7/8n/hf/W/+12xe2Wa/GPH+MEc+eZluZ5l6p42PTNB08g08JfD9zdeqEwAi4jSis9rm81/4Gb7+z34X0Yz5x+++wX/w+S7fG97hnaVVLg01D3ZvEy0ts7sz4mhvlyKTpN6glCGJJix1O8TFPrkYcjTx5GU3dP9MLf/N/+v/jek0+cSv/kU2ts8xGE1Ya7ZRvip6U2kt+qrV/ycYJ/0wYy5bVCUcEM4d7xgNj3nrre/x2ivfJnY5aSrpdTpk0xGqMocMrbuhkyR0m5xKBxhTzs0g6/UG3d4CzVYbFcXIKD4FtLxDiFC8Z8bqmksuAELifeiA+a3f+le0m3U+9YlP4IqiOnSBEDp0aYiwvwXNy8AEK4scEEgtUUqAhOvXr/K//A/+fWpxjJmOQCm0zEEJvFJYp4JZajklJqd/tIfUT1JrLWHKsmLOBgLCwzWyWYt02FtKl9Ef9hkNC4opVXW+pFGr06jXqCUxWkd4DVbUiJOLXH3ik5QUFKYkUsncWOjs/vdoPHk2xnz052d1MIfDAWmaIoTEmGDWceHCpdMWcT9jEX50VugZqcM5x9HR0dzMqNls0m63yfM8xHnGBMamtQ8loXNkVZy+FogKxAznbI1BpykLSz26rTbqDOAnhCBOElqtVvW3ga29v7fHZDJhMpmgVHCbX1lZDW2XwOHhId572u02zjlGoxH+5CS4208ntDudOUPTE7qaFhcXkQQ5qTzPsdZSq9cRUlQGsfYDAKl1lbFj1XKT5znvv3eTJ558MvghwBzEfBR4/37SJLM5NJOhknOAOhAniqLg+PiYKIpoNBokSTKX4oJT/dPQzu8fes/Z8Y9GI/b397HWsrS0xPb2NmtrazQajcfmSz9ZINrsXGc15BAfSgKJCJhfo4eASgEIQavVYjwek+c5CBFIES4niWPiWkKUxOAVtgxFr249xcQeUziSWo39wTFNb6ipmEilpM0mhRTYqEbcreNE+PzheEx/0EfPchd3yrqPouh0fn5IIPXRHDhNU0ajUSgSNBs0GvU5phCMtmKyLMdWTPqZGzkEXEMpNZ8vHzXW/E/CcIXFDTJioZBFwfjGuyxcuwSxpkg0MoppbW6wPDrh7qtvMppm1Jd7uLLAlRYfRWxcuYIaZRycHDCY5Kw3V3jw1luINELZnERK6l4SodDGE/scXx7x/jd/n/byCs8srzIYjtE6ZhInFHnBcqOLareYrC2xd3uH9aRL1gQxPKY9yqmLMB+KKBg1e6eQQqERmIMDpOrTyg5RWczUb9JYXMElMnQcK4GRwddE+dBRKIRAi5gCg/Gestnh8i/9Mlc/8RLlapuTZpOJrSOQSFOSNluYWo1pnhG12qxcv8YwSahdu8wX/v1/j2+c3+Cd/+7vkR/3Sayi8B6NoE1E98olku0tpiPD5pPXGTjP0tYlMiWZ2bBIDxpRkSmZ+3mUlRn4bN2djaLIcc7O18CZ3vRMC3221halQQgFMhSbrbMoH4BSoUQovCDQOkj/WQKJzRmQyuGcwJUl/cGAcz/CvPtQYOqMVhMWx/Cd+RIvQmAaQLYKBKoWod7CApevXGVpeZmkFsAiawzOhUXM4tndPwrfsx5jg3bC4cEhi8ur9IcDdBSdCeTAWoMUkkhL4liEtjrjwEtkHOOEwBkfWuS7LbbObyKFCKzWhzZROa9kB2OrCG9nLqFVi5AI1e711VWm4wnH/T53bt2m8cQVtFIop6jVWvQLeO5nvkBvoc3r3/oadprhgq9voG0D0ku8CP/wPgBM0ym+SEFpSp2ho6DdWpoCFSW4ssA7iy1zpNLIKAluhWVFfTaGqIgwxpKmFmdrWFPQbTeIRUQURbQ6PYb9E+7u94mcpZ3GpFpS6y6zpjRWSEbZbawNgbLEhfvoQ6p6yt2YgVw+MHAemR1nAfd5O/CPmb2S5zlRFHFy0mem5XX58mVqtRqDwYAsy4L2UpyQpkm1cUpGoxFpmtKo1ynysHl3Wx2mkyEQzHE8jmazQVkWQQ9VQJ4PWV7uMp1MSGt14ihUSk5OTlhcXCJJUyaTKWma4Pxp2++s5arKo7hz5w4XL14O8+Ms87sK7GxZoJWg0WryxS9/nka9jpIKF1jtxHGCUhohguzCeDgijgLA4KzH2qAbZpyHSsO30WqyuLQ4byaUMhgMXbl0gY21FW689y744I5JFcwG0oxAilmA6pmhxMFkJ4CuUklwrko8A5NG6wipbTCVcA6cC3PbWUQlgyFcaLEMOpPhb2cMyRBk6NlkC0mBkuA8eVZw6/Y9rj9xDVMaJtMJUgra7Xa4FipmNBwitSJScQhsAKU1CvuRcPL9o8YsyPfeB3fi2qmMhhWhrcw5RzbNGQyG7OzsIqSiXg+GaU8/8wzNRoMsy4JWVxp0q0Mwp8Lz7TzTCug+PDzk+Pi42qwKNjY3ufbkE3S7nRAQE+7vD2qfO/3GKUMBH4pfODja3wdjWF9Y5tY0I0lTbnzvVZKlLk997Hmy4YR6GpMsL7Pc6qJ0RFYWaDTT4RgjHFGScPfW+5w7d464UaO92CFt1vD5MllZUq8KHNMs45//k9/gl/78L/LO229z7vx5rl9/ilarxUuffok3Xn+NJ564xiSb4oFpXtBZWuHg4Ig7t24zOOmTTyaYLEcDtVaDK09cZfvCBRZXVzBasrK9RdJugXw4AJ7x3n46frTxwf3EI+dfgRUyyHt4QbvTglQQ1xNipYnigs9flLxx9Bbv/77glX6KN5av/uJXef29P+TS+hOk/UPWl+q8e2ufizXL+WdfpH1Y4+gP3+Pb3/0279855NqFiOHxHn7tMr3FZY73DxkagxYJRoa1TFj30A1/bAupFygnMRE0W116K9BNPB0JBzqlX6+x1HmatWPP7/+dv8Xv33yLUo2IoxRKy/Z2j7/8Ky/Rak9YbRdsLy+RxDE7uzf57nt73Ng5x7tvG/Zuvc53X/590rUNXtIxZeE5OBpwrb2AnF8/H9yFJQj/4XSjfljWzkdrVHvs2dnjCybjI15/9du8/darRK4k1hYtLZPJCVGlHa5EFPak0iF00EtVVUeTEBbvIU5i0lpKd6FHs9UhTWuIKAIRCoE7O3v81u98jS986Susra4HNtIc2A2a5KG3SRJHMVcuX6bIpkCIWwMAG3bu4E7vkRWLIxSRQ1utlhpnfZWQhDW4125TFgXWOCSSUmgUCuUdvixxpSPPDcNxRre7SNYfsdBbJ6c4E5u4OeN11nLnvQtMExtMI4ejEcPhhLII308SQbdXp5ZKVCSQyuGlZ337E1y+9iWiRpvchE4D7z2u0myH0yTq0fn2KGtrtkfOALfpdDp3+nXeIhGMxkOeffZZVleXH9m7/uzn8qPt5uHYNc55sixjOp3O9WCTJKHRaFQ/d5gZO7jax31VnLbGVMzlh6+fEIJ6q01vYYFYR1VRZdZZchprCCnn+3xcaYfu7+8zHAxJkuQh8HE4Gs2/V5YljvD3B0eH9Pt9Xnv9da5cucK1a9fmoJrzlVSYdWRZhpCSWq2GI5AMqGLKU/ZzCJplOFggFIHr9TrPPPvsXI5ixn79AHD3A679nHFeJfTeh3xzNBoxHA4rJmq9Ilcw199VKpozWYuimOvznmW5lmXJ3bt3GY1GtNttLl0K3ZoLCwsfABE+6mvr42QRTo93LvB0Jh8U89T7bG44Xz+q3D6t1YiTBB1FjPOMYTZlMhnRaNToCsNCqlleWg4+IqbA6Qw31UxKw/LGAjqRWOfYO9hjENVY7HSpaUUkBN7LqvgPtTgmq4zTnLMYE5jaZ8/vbNv9h7keZ4dzjqIsKYuiMscZzdnkcRzj/SkxInSPncaLoYsgaOqeZaz+dHz44XKLNuCUQJU5/t49ynt30d0WcZYgvcZHms3Ll8gPT7h/+xZy1KfeaEHkmFaGcbYW09VLJGkNnxeM2w38+Q3u33yPrpdkhaNrYtpSYP0UHXmismB8/w75wQM67S7nkoSJg73hmKu9BfJ6wtSVTK1Br27iVxqc3CjQpsDaKXJcoNOIKI5RUleyUSU1IfFFRrlzAkOBzCWDeod4fZ1mrFAy7MsOia46hxHh/3UqsKbEKpi2GtTSJoWApF7DlBYrBF2XotYuMOmucm/viE+88CSd6+c5EAInawwXalz9n/5lmlsL/Pp/9v9AHpaQQyRTCiW49NJn0FGbbuzpXG1SuJxSKpzS2LwM7M+quBJLSW4sWgcJHlet2WeZ/bNn0LlTUsJsr5+tvwEDs3gVCEFKRCg8wjjKooC48qLxEqEUUSQRTgCOovSBUCUD8aXIC3Z2H/DM03/8efehwFRZsZVm///Qplz9V1TMo1Al96S1GhcuXGBza4tGs4WodGJmzLPgdC8ZTSZ455HCUBRTYh1jS4PWkmajweHhIYuLiwxPTmjUT1suIh1R2NCAroXACUeR5cGQyStU2uBLv/Qr1HtLuNJiTUmtEkqPVGAYmNIg5pt10K60FWMN5ykrbcFa4rl44TzljbdpNVvcuXOXi5cuEiuBSiSiKxmM4OJzz7O1vck//wd/j/GgjyKw8rzwWClxUiGEwpcFkRDkpqR0Fh2nQQzYBjBVRTHWheDWO4fUETqKccaA0ggV4Z2lLBRSaYqiCAuwb1LaGk5Iaoml2RBgDTu7u2ilSJVCLHTJLEhnaTUW6W0Khpln92CAVieV9ivz6kEw1JhRBUNbpa8Cee89tqr+cuYhgA8yBX4cw3kYDidkWU6j3qDTbTOZTJlMJrRaLU5OTirA0xFpxYMHD0iSiFqtNncJDW02njRJyKZjkIKknnJydEgS10niiJPxgCiKefudt+l2Vomi0CoW2oJihsMRzVabbrfLZDKlNI7kEYA5LAyhreTcuXM0GnW8OK3aBvp2YEworfHOVEwLcD4wqp2jcgSWFbPIc3R0VN0vWQWqoa07CIdWRWAk9TgJi68xCO+YjAY8eHCfJI05t7nB2++/FwojLkgFhPMToAPjQctZC11gkCihgADcej/ThK3YTj5U+RFBW9b7oLeKd2h3GvzOTc3mSc7smrnAmpglV1IhENgq2PdOsrd/GFwzgfX1dU5OjonjiMl0QhTHNJstlJBBsBqJFIJpNqUsS27duvWnPjf/JMbZ4EpWm41SikILguCTrMw2JFqX5HnOyfEhRVHytd/6TVxp5skvhMRcVpIBWqkAeCeaZrNFo9Gg2W6xuBhceWu1eiWzAAiBRJ55H1Hloo8wPvypOaFzM06WxxlDMZnQa7VI0oQH9+5RDiYYY9i4cD4U5nKDShPGZYF0AUDuH/f5F//yX/AzP/MZ6mlCkY2ROiKp1RhnE4oyx5mSuF5jWub0FheJ4yho5yrN9vkLTK3hZ77weZxzXL58mZP+ScW6KTHOICNFlhXkxrC2tMS3f+OfsfP+bfbu3EMZh7SOervJhUsXWN/YIEoTTiZjLl59mt7qCiqJgh7iLHngJx9MfVxi9ye1tj8+aTwt2v7Av0UGPW88wkuUVXjhUEWG7U9xg5LFZh0Ve/Z3UzpO8cnlI/yFAb/1B/fY2RX8f//L/5IXnvsYlzZvsLqxwfJkiZd//Z/ws19+ifIg4fBOxt/92/+Qwiu+dfN16istep1NWrdy5PExJxt1rAoGb8paLBOk1IHp+IPO1QsQnkhqVppr3OrfYn11gSf7lnvTXSayzaevf4l/8B/+JwxuP+DC0iYvfvJn+f/9g79Nx6UM9mPefPsOV841MJOczc06Ni3pLbb43MoKX1z9Ve7c6fK/+xt/jXanzsH+Cd/+57/JV371V8knU8SwxDYFTjq0mR1OpUX/GGrqRzm5/2GHf+Rr7yyDk31e/d7vcvPt11DekChNFHu0Ai1DKcg5R+FMBdRo0rTB++/fYm19g1o9xdrA+G+2WnS6PVqtDkmaIpUGGaSjhFT0lpa5fPWJoHM9Y+yLUyb7KdgbCteXLl8GZzB5DoSCZvhtgZSK6XTIP/2NX+cLX/gCtVqN0pRorSldJdHjQneKtVUrvhOoOMbrGKE1TgiMsWRZST7NybOMG2++yac/9xlef/UGl688U7Wm2odYWLO9OezTQcvVeyhyx3iSU5QF3hVECmppRKdTI05BRhIhNJ3WBk89/wtYvUJuh9jSoYTG+AJEAJOBh/aZswzFs4DMDLSasQhn4Nds3lpryMqST3/6kzSbTeBUV1RUhJA/6/HoMxbHMa1Wi6OjI5wLHUfhd8Sc+TcDZaIoRkcRkdYhh1CKREfIOHmobR0hiLR+qJ38LJNViFMrtvnPvA9dh4S2/na7zWAwCJqQlcwPMNdALcsSW4FpOg7Fh90HDzg5OaHdbs+NmID5nMLaqsCrKY2ZS0ydvS5nW+bnQLgANbvvbqZVKtjZ2aHVatFsNj/APn0UCDwLns3mgnOOyWTMaDTCVIbBM3DbGDNnUoXjo9KCNdWcOmUUZlnG/v4+JycnNBoNzp8/z9bW1kNs1J+ktfUsA/PxMgmz33lk/zsDqM6/dTYfwgcplSjkFePJiKzM2Ds8oJ7VGRYZ+6M+b955l3qtTprWaTc7qEzQrvwY9vd26HQ6nL+wjdQRIivx/THD4z7DLKe3usTUlOTeBkmWPMM6QxxFUIY1cnY8sxb7DxPnPHodhBDElbluq9UijiNqtfDcxHE6Z5DPntPZM+Sce4hx95NA8vgoDhlF+EhTakiMJxqMGNx4m4Vzm+i4ifAxOo1RSZ0nPvYxfBLx/hs3qPuIqF7HS4nLS/YGxywkNXSzTtpscfGZZxGxYP/WTVbWzjG4v08xMmSZI61pjBMkUiCdBZNzcLRD5CTNWpteq0GnE7qQx9Mpcm0BsZAyMAW7wwntyxcwsWTvzn1UMSGe5iwqT0uDl55chnU/kTXiUmAOhgxX+hib01ACjQQr8FJUnTJV/403OG3Q0qAtFLFmjCQVGpHl+LogixztPEW3F7j2S3+Rb9Z+nSf+rV+kXEipW0FpFI6IuMypH0+4LCJsKtnNFZnWHKWw/IlPAAmRd0xShRUKbSExYL1HxuF+uMIHCSKlUFIhOSXmnS3OnmWjzhjaUkriOJ5L3M1MqJQO+J1EIsrQ+a50RCklxgsiD7GUKCnwKqzVUimwoasjkjFlacimP5oR9Q/d5j97fbb1JrwO4I/34QKsbWxw/sIF2t0ustI5sM5XAItDyeA0dnR0EhZgZzDZFNGI2NvfJ63XaLZaZDpimmWBjacUUkjSNEUpxdganIVyNKUQhjKCuFan3V3mY5/+HBcvXkf7BC9ysnyKkIFGL1UIRnQUWqaV8JjShAo2Fm9sSK6NYTwaoaIY4R1xpNjf3cFYy3JviXq9Ds4QiZI0Eui0TaPR4st/4d/gn/7Dv8/k+IhUAMJhKp6n8KFyrKQCb4POg88Qqnot1bxSK6M4gJomwpgSoSKEDv+KsqiAEI3JJ2TTPkXRJjNdCuvptBpEccTe3Ts82NlloduFeo0Hx32azQZpnICPiNurrF7UHA+nDPpjosGESHuENVW1zoeEVQRNMB8Q1oeArxl78M96pGmdg4O7NOp1Ll26RJIkjEZDPG4ehE2nGQsLDRqNGnGkmVQmPkqGCkm9VgutGEpTq9fIiyIALlUwGscxvajH+3f2+fYffpenrr9AtxNTFiU6iiuNSsWgP2BxaYko0lhbMmPKnt2a8zynXkvpJr0Ajs5ZFq4CEyonTwGlcUH83ll0xYgpjUVHujo3wXg6Pq0Ai0obU+mKFSgQasaOFQSjC4c1JZNRn93799jf22VlfY39/b3AnCW0h4t5MqMAUbFQAoNxlsAE8Dckb7Ok0FeUfHxlVOcAAuAxW+ydNwgV2qSs95VJQgW2Eliqs8+ZdWfO15zZuXjJ/fsPOB4Mefb6VQZHB7iqVWwynZLnGQu9Hl7AZDxB6YhaWgu6l8awtPjH10n5cYxH2xdnwdfsax1HqLhGnBQkaU5ayynzjDwvMEWBKctKX7kCqqsK4ezv4zgmrdXC3G/UabZaNJpNkjQhiYP8SFikzzhe+zMg6il9/QPHLaja7ap5g7Pk2RThHGmS0Gg2GPe6vP6t7/JU9HH6Jycc391FpzFeSTqNVtDWm05Jk5Qv/ewXaC10kErQqNfZ3d1hYWWZSGtqUmEmGZktQRCeDeu4d+sum+cvsrG5Rdppcbh/wEK7xzd+7xs8ef06k8mIq9euIZUkn+SMs5yFpRWMc/zB177BcO+ICIF1hma7xYXLF1lcXyNKE8ZlztXrV9k8f556sxmAEw9n4dOfZCAVTt2y4QcXyT4swPonsVd4whpHZZYovEQ7DTLn6HCf9998jyaKtV6byWDM4fGEduy4sJWwcr3Jx56/zLe/cZe/89v3+dr3/oDv3m5xafs8RebJ+yP+F9vXGA4G/Nb9t3n5/k2uLF7i6a0nsF6wUOzRm77F6ljx5lvrpKufIBeGxEkiBO6RY31cwuyEI1MgvSCNE55+6gmO7t0hTo5YVRHu8JibN26SLi7zyc0L/KXPfZmnnnuCv/Ub/xivGxxOBN/43Qd80+b4eMrf/rU7LLY8Lz57njvvv8yNm19nYhc42Dvg/NZzXH9ui83FFVYySXTzkLK+hmx2KnM4hxMC7cD+pE/WP2LM7oOrCsVlkXG4f4933nqZ999+A+ly0ghiFZIX78C4GUkAjPVIGeEdlNZx4eIlOt0utVqNKIlJkpQojomTlDhJQ+FbKbwM7DXnPUmtxouffDHsy1WlxVXaqKJ67Z0F4eZFVT9nbQT324BxhX2z3qiztrbGqGIGAhhrkJHHW4GwApyo9M4lMoqRUYqTCk2BsyXj6YQ8K5iMB9y9+z4XL53n9Tde58lnP86oOAVRz64FMyB1xipx1mDLkskk5/hkSFZkeFESxYJut0mzVUNFAlREs32RJ5/9POgW1o3JyylapJTWIkVIBr9fEedsQfpsouW9ZzCoWvqlCBJJzjAaD1hfX2N7+zzNZuMDYFoYHz3HdCklnU6b8XiEMSVJElemVCXG2Pn9Duch5y2QWuvAKAVEVXBVStFsNlleXiaJY+bVqtm1gBDXPwYgm43ZtW61WhRFQZYFeaHZnJsBnKU18/eK45hGo8ETTzzB1atX523MEIDUwpQID820hoe5XMDjxqPt1GfHKbgeXi8vLxNF0fz7Z41xz0rVzUDmGTDtnCXLsnknW1JJHQRGoZ+bnAkhHmLBhr+X1XuqYEJ0dMTBwQH1ep2trS02NjbY3Nyk3W4/rJP5mH3zJ4GN+IOB4IeJI7NAcbZ+zNeMmQ4uVDGbQkhPNp0Sa0UzianVIqTKkYnFS8NUjJkWCcP+gAaOhC3KYp3psKTbiCB3ZMNjUhTSQqxTlnoN6s0Wk/4RxyfHWBkykMJIrInDXPEzrkC4J7O5+kfdi0fXJFOxwdNarSrogLWuMiELMgIzEpIXYJxFK/1QkeAswP+4+f7T8f2HSzRlTWOFJ04iKDLym7dxd+9TdLsQJdQzga0p6La58sJz1FXCnRvvYuSURq2OxNNd6OCs4t69+2wtrtBpLCIxXLxynVavFyTrugnHRY47OqRhBW0p6CiFN4ZShrVsNDpEmTFxq81qq8vElsiOYhwXyHFBb2UZubhMP5twLyu4emELl5XsHB4ysFMiYdEiJpYRCI1JYoyVyCiFWg0RJ0gnkB6cnuXEIvjfYAPm5AWJr8wGU0niBanTTIUll5ZcFJikxhNf/BKXP/Vx0uUWUyHQPvhoKAfRyZSj773BVmk4kZ7pco+f/eovI69tkmyuk0/AxcHcTXiBMI7ISVQUIbWkxOB8wIuUCjJ9coYnnYkpvH9YE31GzAoFDkVZGvK8wNogsRJJhVQRMoSy4ASlhAKH9R6PRFg3X+N1pNBCk9tAUHM+yGae7Q7444wPBabOTEzEQx/mTwvqvjrhqj03rde5ePEyK6vrgSLtAygS4ESJFB4vJQiFNQZfgUeiyJC1OgbP/t4DkljT67Qr9/iw78+0kIwxxFJgyxKlYmStwdal8zz/8Rc5t3WeJKkTVa5hRilkpSU5Y3PhIVIKZyuNydJwcnzE/v4e66urRJHiYH+fPC84d+48RVkQR4rlpQUaaZ2TvT3uZRn9cZ92u8nayjpeSAa5QTU7fObzX+Jrv/EblKMBka4o2M6Btzgn8VKBFtiKUj5vW1EV3dnZAFSWJVYphNLoKEHoOABkWuNKBZHGe0cpoMwL8tJSFhbhDHbU58brr1JLIgYnIXiRkcJ4R5KklL5BogRxvcuTz7+IjhKm5R9Q2geY6QSsIVR6w7GJObxauYSe+Xe2DechxtqPcRwcHbO8usZir0tehCp+q9WmKDNO+n2arRb37t2jUdRpNFKiSDOdTmnUapgysDnG4zFFXlBPHUWZ02y12d/fI88yRD3Ml4IMLwS3797l/Vu3ePrpzlz7ZhZAjsfjYD7RaHJcuX7q2QJRbY6vvvoan/nMS6FSOWemzGy/QqtfYAUIvA7AUGhZC2yBsjQYE9qvysJgTElar1MWOVJHeBSeoGkpEZUmVrVgOYctCgYnRwwHfYo8q5xJIyZVO1luLZFS4fmrAqJQeZbgFdYEbShVMfFm2lFCBO1hD6fVXe/xlXlHcDslPPdaB6hX6dD+T9Bdk2J2rDMNv1kgHS6TkBKpNZGKcM5jreDmrTtcv3aJ8XiEs5bDw0NWVlbon5wECRHrcM6y2F1kNJogKj3aduujq5n66DgbZIWNISKyNdKawZoyAKdlefq1NXgXSjmummNKyblsgtaaKIpIkjjoDcfJPDk9Ba0hAKgzbmmVfFVxtDg9uIeC67NFlpnWspSKzFq8NUgB5TSjFad84ee+jMNz7ckn2Gs0sUmEKQq++6++xrM/8xJxp4kpSqJayng0Zndnh6tXr7C6sUmWTRgORohGkyhOaNRbMB5hvWf37n16nUVG/RE3336P7uIi77x9g8986iV2dna4cuUK9+7d5/r1JyhdiQNavS7t3iLf+dZ3OdrZwUymFEVOt9PiytWrrK6vkTTrZM5w5fpTnL92lVYvyBD4U0GD03v2pzQXflzjcQnfo/IOP+pa//DfP14i5oPJfcCbEG5e6HMSjim40z/GiIh+3uLOfgPdK+iul2RK8vabJ/Q26iyeX+Dn/s2n+a3ffZuFZo+j+/dY7S7y7/6Fr+L6A0SzyXtvv0Y/2+GLP/fLvHRlGey7FHvfIG1/h0O3SGPyb1Lrj7FoMulJran2yA/B7nQeowSJcjQXlrn6wjO8c3eP+y+/xlTe4Q+LKc/+8l/kxjdfxU6hOCr58ld/lVq7wftvvMzOO++jbCjIHu8UHO0L3r73NlKAzC1Fdp9ECNJaTGtxgWc+8ylaU0N0PKEUPsRHzmErANgD0nmceNzRfzQKpn8SY1b8y/MRt99/k72d29x6900wOVqDkg6lLKWVWGfx1oEIOulSJ0RRQq3WoNvtsbi0QJKmJEkNoSRxnCKVQlamNJ5gNiWkCh1YM83AeU/XbJ0MQGlYZx3WlghcMHjEVaBXkOQJ3UGhfT+wTR0vvvhJ8jynLMqwZgvAWbRQaBET6RitBVJFOAFxonECstEIZ8bs3nkPW5Z897uv89kvfIXBoI+YlDQ6XfKsPNMtcjoPZsSIWVHSWYMpDQcHxwyG4wAYaIgk9Hot4igljlJarfM89fzPo9IWZRlib18W+DSq5Ad0ZeTmP3DfTskbD5M5siyb61e6yrG3NAV5nrO1tcXFixf+CHOfP/tV+uz6Nju/er1Oo9FgOBw+XECt9u3Tbj9x6nZfdQsVRTEvlq6srLC+vj4HXx8PKD9+HX/0d7XWdLvduaN9t9tFVC3uCBGkyiodPCFExciLMaUhiiMEUFaGo2ktDfGjY54LnW3tfPTY5vf9zHHxgX0hxOBCiLlu7s2bN9nc3JwDv2fjm1kSXRQ5g0Gf6XSKEIJGoxEkOoTAWledT/gMVz2Srsp5Z3tWWZbs7+9zdHSE1pqNjQ02NjZYW1uj1+udAb9/8sej+8EpU9194Huna9ZpnvjoP2PMnHkWcp2EZq3DytIycd3T6IBMMko7Amkxpo/zKaXrY4qc8+euIpFEzqAMRDL4hqRdhZAWqwTaJIhpxHQ8JFIa7ySmyo3sbA5XYzZXZuD/hwFVz+a+Sqmq88FRlobxeIKtQB1dsU5lpFEz1vgZ0P+x2rI/HR9qSCwkMrS4C4lQFnl0Qvb2e/irFzCNJpkB72My72h32lz5+AuUUnLjxg1MJKmnCVLH+DJhbes8wsPtd27RUpLNi+eRsSBd7OK0ZO8kB6XZqLfY2TugkJJuI8Jrj/MG5cCfjMiLgvujE5Jmje1Oj8PhCDGxtJtNWlJTTg1xrYlY2cJmU6YKpmUfsj7drGA18zTbKceiZJTUiLoLyLQRzJscQR9UzHpWAuFNSJBW4rxCeInEEWkPtsRIifOStFBIBw5Dmsb4aIHSiRCfAlZ5ImmxsmA06hMNhuhGnXS5xYXPf4JyawWfJCjvKYRHOYcyEqElzni8Eggl0V5glKoIUhU/h0DSObuuCjEzoppnmkCIkYwJBaw8D93YoQgSJIsUApQKZE1n0dYGIppzSOmJiXCEdVw5iaqwCGdtRTf70Z6zDwemnm0TmZ+Yn5+nIwSHUgXA9cLFi2xfOE+n250HlFRaoXiBEBXoggoC0CIAiMpZJoM+Kq3hbMloOAiLrPPEaUqSpOCDKPl0OsGOpmxeuMDa+YtsXnuC9uISaZySSkmsKsaKEGihiRsJ1pq5G1gSx+A8D/b2uPX+Lay1dNotzm1u0qjXSNKEXrc7N6yaTBTJxjrGGpSXxEpRqyesbywzHg8YnhzjhaLwEEcRje4iH/vUS3zza79JUU7RAoQKGhC2ankVXgW2lzHB9Emp4HI9F6AGJzXSSXCWwhqkDgxIYVRwXM+r9nKtKS3gFRSGqJxwMDzGZEO8ajIZe0pr0VFEnhc0my2sMTRqNQoJUkRsP/UCaW+ZP/z6b/P+e+9ipmOwJbJicZx2/82hvtO5AB/YHH/cQcP+4QGlMSwvLeEqraMsy4ji0JojhAhs0zxn/2Cfst0mjiLyPJvrR9ZrDbIs5/j4GJUojk6OaHc7DPuCPM8ZjUbohqIwjsJYvvvy97j2xHWkDPPEV6CREpKT4xMWFnocHB9hbQnE8+uiZMTHP/YJytKc2ayrSg22YgyF6+5dYATkZY7UlVQEAmss08mQJI4w1pIkoaVLRQlB0jQADJGOkTK4/HlbIrCMRyMO9h9gyhKTF4xGAwbDAaPRiEa9EYyqpMDaEq3UvLo6q8j7mcswAuFlZZJWJTeIIC9w1s1ceLSKwu87h1QaVxUMnLdzMyMAgZ4HzCr8aWDuENhBHoJsiA9sS5A4L3jjzbd47plrLDaa4C0qSeZBWr1WZzKeoJXm+PiYLMvpdDo0GnUmk8mPdZ7+KONsYjNbk5WGOInwNplf0zlrfMaInktxBMmFmRzAKTNjBpqdbX08dW+em7bMjoPZ8v94hshDxwpB0y/PGY/HTEcjRoMBpiyYZhndXoeGlXz793+Xz3zpC7z68vfoLi3y8S//DA0Hu7v3udi5RikgbdZJhMbkBTt7e6ysrOAN1NMG+ycnNOpNYgXWeqbTHI/kd37nd/n5r34VX1qmJ0NWl5ex1vLn/tzPcffuPa5cuUJpDSUWrxSNVpesMHzvG7/P4b0dbJHTXehw/foTrCwsISPFuMi5/twznL9+jfbyEjqOEZUxi0d80DD8g5fqJ2Y8Djj9kwrwZ/Ps4fd7eK59388UEIAmiZVh3SyEYNRo8vbJAbTqfO3tQ+pS0GlJ2rcUwpYc7N7EFXcZ5AW2s4CMauzaO/hJzlPnJB+LO5S1lNsn93ntlT/kL/zyZ/nFX0gZnfyntOhT44AoP8A3z3P9sy9wmAtiZ/Be4iqpFj8v/jweEJIetIGyOtVIpJy/dInLV97n1a/9Q7ZfhCI1fPONN2lcuMztacnBv/znpEWfb/6L32UyGRApQT2OcSR4bVlY3KNeh+N9TVk2KZOUqBZxPDiisbPP737n9/nylRdY2erxrckuz8seHSfxOKQ7cw/848HgD96nn4zxAeDIG/AZr/zh19Ey587NVzHZkCRSaB0koyZlSVEG+Y/QQSFotDq02z16iyt0uz1azSZpGjQ5lY6RUiOkDsXuSjdZKklRlhzsPGBhYSGwJr2ds+KqA6ziwdB9YYqCv/93/1uuXr3Mcy88N5dIEQSjMOd8BWL6UCzzlTutV8RxNP/cMARSBfMrrRThIy3ejhj3j8gmU95481Wk9vQWFvn0Sy+xezvEwy888wxZlmOcIDrDSn34X0iAnPOY0jIc9Dk6PqzkfgITt9mq0e21SZMOK8sXeer5r6CiJQo3Ii+mUEakcUJeThFo4IOmK2dZiTMG5mxdmrWbG1OidGCjTqdTGo0Gly9fZmNjHSm/v5HVR3nMALoZEWQ2zuZjM9mJWUskhLmeJAlJkrC+vk63232ssc5jn2fx8M7+uHUsGOlElGXJ8fExcRxz0u9jnWNxeekhUgUEY59Eh9+3lalpu9UKpQFjyfMi3Ff9QYbe7EDOAqkI8RDY9LjWfQjAr/ee7e3tObhbVFqWZyUKgmxBH+99MKqtQE/nfFV8jj6Q24RiQojR8zzn4OAgSInFMZubm2xsbLC+vk6n05mzZP91HKfXvyr4PLQ+hPj+0bVjxkh99HUcx9RqtZCHNOsspS0imbDYWqRe0yAmxC2L9RlRLCCq0T8qEFLhRDAGNaZAIri3u8PmpSsQBTC8MIZGt82qgpM3D5HSIpUiryQlHn0Szs75uYbwH3EP527k4lR3eFb8mM23KI5xzlFWEmtSCIhdkC87E3fPCib/us6bP60xHZ4EY2hjEUKTRDEmy5i8d5vefp9hrUWhPXEpiRE4LE5HXHvmeaJ6kxtvvI4fT+nUahRaIFHY0tLeXEGYkr2jY7LxkK3tbbwQTKVg5CTthXVuDydEUcRkOiYRlrZWpFKF93AyFJHGQ2R/zIps0NQJfRx6MsTkY8RqD7otjt7rM9qf8NxTT3Iy2sdOckb3+4iDMWkn5uTKAp3tLeLOIlpEoftbWGyF08x8b2TVuekRQb7AeVSF3eUCICItg7m4UxZnSzQRwobOViscKLDOoNspx8KTJg0yIVm9dhm90mWqFWJqkCJC1RRxZpnkJWiFkBotZDA2deHY7KyoVoGpj8aap2v4aZZprQ1GwlXBY9ZhYIzBxMHvyAmP1x68JSpKsoNjWp0W48hjrAsG3X5maC1QZ+SbhHMo8WNgps4WgbOByEMLDSGwkSq4R1+5epWlpeWwaMxR8lklW+CFxaPxHtQZFor2jsKWeGpMRkOSyvnOS4k1FqstcRQ0RYqipNQ1tq4+yfXnngtaq4VBUSV/WpE7HwDeigk3CzhCDuZ5++23efvGWywtLbK6HDRsGo0GzhqmkwmHh4dA0AmSUpIVOd1uG2c9yklqMsZbj7cQJRGR1Gw1W2RIJnFMr9VASct3fu9r+CJDmGrzFxLjLcL6OW1Zq2AWgAPhHVQarkLrAJ46ByrCeShd2ATELGCWirLQyFhS2BGiLBmXYx7cvUm33WQ8hqTRweV5YAxaS6wUkfS4siBJgt6SQNJeO8cX//z/n73/DpIty+87sc8557r0meXdM1XPv/a+Z3osZjB+AAxIkFxS4nK1jFVQVARjFaFQaP9QrEzoD21ICuyulhvixopaksISWDpgAMzA9GBsT8+0993Pu/ImK+115xz9cW5m1Xv9eqYxAwLjTvTrV68qszLvzXPP/Z3v72s+x3e+9XXefPVlTNx3sjI3vR0YjgP0Rt3hw8XWqAj6q9h0zc7O4XmKNMsIPI9qvUZ3v4OQsLu7ixWwuLiEydNxumKj2UAJQb8/wFqL77swp1gK4jwh1c5D11OKWCc0mk0ILJ1rm/iBz1tvvcn169c5dfLseIM08rvp93pMTkxQCkPSJCmk5bkLjzLGFWjFUjJKqccarCkKRjtCDt2S4qmC6SIkUkh2t7fZ39+n2awhC0mKwXmsxgMHEIcqcNJ56xYZnaf09tusr6+yt7uD1k4utrG5xa21NVrTU5w+e5aX33gdP4ocQ0SbsSfrYcmUUgXEZgoT+aJhgBVID1fw4oLcKLq/1roEV6zzPc4yB9a7Vda4DpZyKYZGm7FBdWEQO2ZECk8hKTagxiUGrm9scfnSZeYefgCjodvrkeU5SZIw6PawFsqVCsM0cRJArfG8gMD/QYyVn7xxuLhyrHYHWgupsPaQmf2hv53LqetpjQHvsdnCYXD0cHFcwFoFuHLAQGWsUhCHnnf4NcePQ4xZqclwyNbGBpOtJuWZGfI8Y9DrEkURnoH7zp4n8Hw+9suf4HvPfZ/tnW2uvf0Oy2dOkejC89pYEpsxs7hAe2ePzt4+5WqZUhTg+QFBVCZNY3xp6A+HlMtVJudmGeqcoFrh1toqjck6t27dpFZrsbmxxcLCPLv7O4SViLBUQSqPzt4ez3/nu+RxzNzsDMfPnGB6aortW2s0pya554mHOX3fvTTn5vDD0qFAjKI4cFrVn8lxt7X98Jw5PH7YRuDuG5S7/f53P9eoIgjNKmeRQo41Ade3O2zs7hJ6JZIUhmnM7n4IGIxNUZTwkARhCENDhKE30BytNfhPPvNF0qTD67fe5OzpZT731H382l+/n1LyNRA9yAcQprTTeeaXPs1QVChPldDa4jkhE9hRgMl7A5LagpUWlYO2AShNw6/Sak5w5pFZcv06XjfhgbkaX/nTL/Pc1WvIdEASejSbE0zNzZKhSPZSZup1TDDgP/n1B5mdbLO5uc/qtT7fenmfN28p9jY32NvcYy7OeXjlDMw2efPCa9x3z1mstiiK4htIlcUz7x8s/+kEVxO++bUvI8lp72yQDbuEgWOjGgRJqonjlDQpPOul4uy58/hBxOTMPJNTM/h+gJIS3/cck0J5KOVjhedqI+Ga3QhLmgx4+k/+mJMnT/LEk0+ANa4iNoXfLwACg/PykhKOHF1ibX2VB8T9WONUHhSkL2vNWJ2ltS2aYd7YI1RKx7pQygMpsUriBT5ap8Rxn63VqwTkfP+Zb3L+wY+zcOQUUUnQ6+5z5dJV7rvnARdKhFewQ7hNknunAsla55WexK7ZnKQDsjzHWEXgRTSbder1GqdO3cvpM0+QBnNkWYLOMvLMc1ZIeYpnApAaI1OMHt2xDsbo9Uag4HA4pFwuj6XmWufoAr+rVp20vF6vjZ//0wJKHL6m8jyn3+8X/nB2LCkfK+woGHEFgHM4RKlWqzE3NzcO7LgTcHwvdup7jTsbpSPrgI2NDTrdLru7u6yur3Pf/fdRr9edWsCYgmxtSQ4lmlOENJmCgTxKuJ9dmH/XObgbY/fOc3XnMY1BreJ7hyX/a2trLCwsEATBGAxO0wTf9yiVSmPCjTvHt/v2jv4esbF7vR57ey5gq1qtsrKywsLCArOzszSbzdsA25+Hcfv9wI6B1MNrx50S/ztBy2q16gJjTc5E3ScZ9vFFhDJlPK8OSYAnFaWohsiGiDDBSoEWQ/JcU4nKhGGLhVbTrRFhBaQiH8ZcfusSzXoN0pycHOH7WN/twUYKvNFnPLrGDltBvNcxH2ZN53mOkII4SfAK9WueuwaPy9VwwVOlShlPOHAnSzNSmyKlJAzDMYs+H2W4/GK873H1tdfZXr3FohXY1OD5Hl4Y0Nnawb56mdrEFO2GJMgkkVYk1uKHZUphxPLZs5QqFS6/9BLdTo9ys4oMJHZoyBohmfAJdyx0YvpIhtsdlqePMJifQ+6n9EsB08vH2L12jXKWkiYJwzwnUJKSllS1wGYZibIM9BCda5TOyPIhZZtR82sk/R556NNYPE41muHyzR3mjp9mOJXT21xHkuCfOk40N0sUVVBWkGMwnkULi7LSgYYAFnJlxhs5q3AYjpEYJfCsRUsPi3ZgqpLYRKOsRAvwrITcKag0glupJpaC5dOnefRzn4NyGWEs0ipybTFSo6zFBpJYGMraMVyNMGMcI0Ah5IHKFA4DqLyLS+HuZy6T5cBP1TVIEDCkUE0LMNIQeNBZX+fNb3yXB554lGChydA35CgXUiUFUgl83yPLLUkSA5ZKpfRjzbv3F0AlnTQUe9ChsUU3XRdnRABRGHFseZmZ+QW8qESO+wDdjb5g7RiNEaBtDqpArAGD+7CkzhDxAIOl3d5DSkGjNYGwmjxzniNeGAGCsNwAAnRmCUoBmbGQZ1ghSDLwSiFZnhMIQZalRKWIZBjjewHPv/gCW5ubLC0tEYYBVlpee/1VauUKjYYrwPY7+0zOTLO9s8P83By+9NB5URQgCMKAarVCEAXUmy2uX71Kp9PBq5QJfI9ch6ycv4djR+ahs8PW2k2u3rrFxl6PLNVYEyOMS0d3lpPFhHNkO4R0IJW1DkhSXo5QCuH5gI/JLcrzMDpHCQumg+cHWAJubbaR1jg030udbNoPMTYEKRjEMdq659cQhDgQKzeaWm2CJz/6STJtuPTGK9hhz5kq4/wnzIgh4bDVcRLi7QXQX74HVTkK0HnO2upNlhaX8DxJo9XE8xTLwYpjbwYBKMX01DRp4mQ9C/MLDIYxSZo5H1JPUamViXdiJhstNIYsSSlVSyhlCaISQvj4YZlOu8P3nnuGI0vzVEpNlPDJrCbLU5Qv2d9vU29U2NrcQVcbdDodSuUyE5OBAwKEQCBR0isaDqZgAVuXY3OIweL5PmmmUYFbiPY7PVY31jl97jRhEGAEUEj7pR+ilCTwBNZkWKuJk5huZ5/O/h5CCqanJkizDKSiO0zIbtxgf2+XsyeXmZueZGtv3wVXCVcQSyEci0YpGBfWblGSxjsoti0oadFCjw35jbUYBJ43kr7ogjVQZH9aU3ipjgA+gR/4hSevHTcO5KGiRRE4xgm4hkKm2d0foIWgNxgQlSqsr69Tr1ULdrwDU+MspdWaoNvrEsfDnypm6t39tQAOQilGY7wy38mSPMQ8cx7I7nsG3oVajR41/j0jINYWrONDn9joniCE+67Gee4Ia0m7XbwsH29ukyQBJSHTSN/Sw/DCc89xJr6f48ePct8D91Oamqb2gOTpr/wRT33ik0TTPlqA73n0+12kL3jj+Vd48MEHMaWQNNfYOB2rB7CK1VtrnDl3Fp3n9Pp9Gs0G2xu7nDt7js3dbR545D7QmlpUJbbglytYKfjW00+zt7fL7PQUZ06eYHZ6hu6wz9yxRc49+CDL95ynMTODFwbFGujOxeiz+FkCUt8PYPZej7n7fL39ez+KisGtIQZjQRmDhyQTYCRsXrvE9s1r2DQjsj7SOh9yXYTaWKNJRMYgNXjKoyJCJkXA//5v/z3mmvN8f/c10to79DtX+Z/9yiLJzrNovUoahOS6BzbgRv8RauX7aUiJhyKXebEBD7GF7+PdjvPg+AVYD0uKlCmJUCACpmdnWL2xydJCwOzcPI2Ty/yL//ofU6nWkdOTNBsNBr0e5TCiUWmwl+2SmgTPb/LMGxl/61eOMBn1aNXLBGqKfrvLyRPH+O5rL7Dx6ivEm2uk000uXLlK1h+gfEUuLJZCJSNcMOgPPPc/wQCqE6ZZrPWKDYXFigyLQloJOuXia9/kyhvf5N57H+DG3hq+ki4N1/RJ0ow4VgxjSxz3yLKck6fO0O/n9NY3OHfuQcIgdJ6UnmuAKikR0gfpOyDVdRKLNcCjWp3iP/g7fw8X65kjhPMxczdst06BRAiNFc7i5vHHP0iuU7I0H3tOiwJMzbUm1xqBxJOuTpDK+ZIL6Ts7AukRSElmMoTM6HW2uX7lIhs3byCN4ciRozzy1McIS5O89urrLC3NUY5aPPmBUyAVSZKRZmlxJzausV8wXUyB6hrjGCLaGLI8I81T4iQmjRNMloBJCWsBx06vcP6BD3Ji5XG0LWPyDKMz4mFGFFXI8uwQA9EVlqN7zNiTHQqQy6C1A2H6fQemOvVJhrEGow1HjhxhZeX4eE68nzXoJ3VorcnS1K1fuSZJU3Sek2uNlALP8/F958M42qB6nsfk5ORtsvbROAw6vtc5uK05+h6MzzHDXikazSbdQZ8cQ6VRI4hCcqNdKKmxhf2CJYxGAU4jRZKrQrQ2zMzPOZm/ZZzujii8jTkASMeMVQ7A3bHs3/GMxqDY3Ri1UkpOnDgBwP7+Pv1+v/CBreP7XkH4sYeYqLooiUYsbDMGuLe2NhkMBpRKJVZWVjh27BiLi4tUq1UX1nmX8363c/qzNsbNFnu3BsztLNXROGy5EAQBYRBh+gPC3U3C7WtM7PqUyxVkqYUoz+BVZ1C5wauFZJ7H9t4WaRQQ91JKC8vs9mNQhs7eBiv14ygZOB9qY+l2e8RJilLOxsXTPkFUsOQKD9ziTY1tNUZ/24LVNyYXIRCiqHaloNPpsrm5zbA/JAh8YiUJA3dfUMov5kaAMZYsychVRuSHZJkmrJSplisEXoCywHaX9sYW5ZV5gmbRFLpLOOQvxu1DVkrUG5Ok1zdJhCXKQZUUKonpvv4K9YVZonuWMREkKiK3IZ1ME0tLGEgWjyzQCENefvsN2purlIOQsl9C+ZIsySiVy9QWywyUYS8eEtoq/W7MXK3FyccfRxpLvtXGa1QYbm2h2x1MlhIxQFmfUiCJlWUgMqS0REYjMlBaE8U5USqolCrErQqp0iSBIp9p0vc9xFwTyFEnzyCmZvDDAM8XThVqBC7VxA2LKzEEoGwRKl78wGCR1u27jShMnmwR/KoKYewBkwapBUJ4LD75OBcrAY/8rb9FeOwYQxWiMAhlCcgxUpF7gkCEWKORnrP/tFY4YqUAFXru/RYMywM15EiJMFK4uhb/gX2GcApVpTBpSp5opFRo7SFsgrLG4VxWsrl6nZ23X+GWZ1ic+jix56FNSkmC0hIlJEoF+IGzx5RaMNGY+LHm3ftkpjr6uTp0Q7UFXRatC7NzjyNHjrC8skK1Vic3Bms0ShTF0Ch8yhTm5EJgivRDtwl3EjmJwWYxMgjJ04Rhv0e1UsWTCqQiTWIi5VGuVukPMuqVCtVqFW0tfhAQ9/p4vocwhjxNSeIEL4wKf0TX3X3rrbe4du06R48s4Xs+WrubYxAETE9PEw/7tCYn6A8HhGFIEidkWUan08EKSa/Xo1QqIULJfqeLUpJBr0e9XidLM3qdNmFYolStkqce1WadheppRDZgv7vPzbUtrl65wfXrl9je2cckrjh2N2Ani5C2kEoXJv+ykAAIGE9Ca6zznxSCTOR4nqQa+XT3tkCA7wdgLWk8IFESFbrggtg4+DNOEpcGniQ06nUa9RrWGHY7farlEh/75GfwhOXiqy9ikyHkeTH5GcsYxfjzO7ghjqXBf8nFQikMia2lUQQ/GWPIsoxWq4ULbrBsrm8yNTnBcBBjTE6lUqXbc34mTjKtiEolarUySnkkSYryFEMpyfOMuZkp9gvvGyMkYbnMq6+/xuMPP8zK8TP4vgs+SIpOd6fTYWbOBRxleUaz1SIo5OdSOaq5FC6p3hxCYEaJvaNujFKOfSWVh7GuyF46cpSwEuEFAcZa5xkiXNK9C9ECazVGp3S7HdrtNmApVSponSNMjhdaVFRBC8nNm9e5fu0Ss1MP8/jDD/LvvvqnKKGcDL8oNFwKnwPUtdbkOndzoTC8HjO/C4+kIAgcGxfnhWW0xlMKCssLz/OxArQxKFGEcBkDxv0ujQP8HUhri/A4UdwoCsatKtL7jOXKlesYPki316NSreEXXrPVag3pKXbbe8RxTLfbodPp0Gg0foiX2k/DGG1w4E4U70726V3H6B77g0ASh6OP97u3MVoPYYfuMW5ro6zzYBymCVmWUPIVzaAKucH4AX2dMYxjXn3hJT7yhc/w6V/5PFdXN9BScPGl12nOzbJ88hgf+fQnMKUSq5euEucZ86eOI63BL4ecPnWSzvYONlRITzGMh1QqJbIkJghDsjxjv71PuVwmTVPm5uZZW3X3ohs3rzO7NMuVt68wMzmLV60ggxBh4MKrrzHVqHP29GkWFxcZDvvUW03OPHA/K/ecozY5hfQCR/Ut+OWjeSnuOG8/7ePHAVPvHO+1wf3zDgFIW2xwldtEp3nOH371K/zh7/4eeRqDgqHJxo2a6eNHOHbiBMEw59Irb9AZ9JBSYXJNo1WmUWkSR1U6XkzL9xmsbzAMbjA976F0mXSvS5pU+Z0/ucl+dpUPfGgSvywRQhFmGk9DrPL3vPfd/j3LKHBSF/JEjxQVeLx8vcdEs0pVW+xel//wH/5Dlmfmuf7G2wxLin/1L3+HWxtbXA88fOU810R7A5E2MHkbr5Rx7Z2brF+14M/zoY9+krWtLT72q59jbmqSvVs3KeVDXnrlZT7y5AeQxnN2PjLBcz3TH/oZwk8oqGoPv33LCGJxzE7L6rW3efabf4zNeuxt3MTGGVJpdJaRppZBnDNM9sm1Ic4yJiZnOHZihWNHj1MKSwSeQolC8SQVUvpIGWALiT8F6WDE+x8B1Mr3wUqsHqkpnOmie1zCyFPQGuk25wYnXTPOMsiiMSbB2JQ8LV5fAUKhvBBQSCSeL8FmbO2s4kl4+qt/zD3nz3Hp0kWWT53k8ac+wn6nQ2d3j1e/+wIf/PDHue/e89QbTdIkJyvASisO1AdWa1c/Fc7ERlsM7t6f5xnGWJI4Yb/bo9Prk8SaeJggZM5DD3+AD3/4i8wtnCBJJdpIrMmIhzG+H2BMDkXAqRUjT/lReNKdAN6B3K9UKjnVSuGR2h/0sNby+OOPFUzM28HCnybg6rbry01hKI59BPZkaTpmJ+d5Tpa7OszzPM6dO8fS0uIPrGt+0Pm487q+G6B6mAFardUYpDEzs7PjQKZyqUSW5ejcNZnq9bp73wXDFg7Zx1jrNv1SFiFAbt9z7fp1FubnCf3AVRTm4DM1o0T0wmuyeJa72u3tkvLD79vzPJLE2XhprW8Ll8qybPz1iCgw+nrkTdvpdNjY2KDX69FqtVheXmZxcZH5+XkmJiZuA1HfE6z+KZqL72fc7X4wui4PM5lHVgDvq57wJLFNCHOBuHmLqZIlKksSH/BCPBUS+CF+fZqkMomqT1K/7xGYKeNZhZ8OHSPZC0lzS+C70GffD5BKogtVgLUFVJ+AyS3a02OGvy2IYNZaqtVqocYb1XtyTCBwno+CG9dvcOHCJW7cuMX09DT1ep1arYqnZLGPd0NrDVY66bRw94tWs4mIQpTwEEVexN76Ful+DzUYEjSqPxO15F/GCGsVVKXEMJD0sCiTU7GK0BPkezvsvfgKcwsztGclxg+J+pqs4jEgR2YWT0B1booHpj7I6ltvcfXiRRgmNIMqoeeRVS1JkuPllsUTp0hkTm97i75fhSgk9EKOnT2HVobrvX1mK3P4QrO6eYOkn1DJDJENKIU+Uhu0zVGBwpMS0+4iukO8aoQ/12RYarB0dhrb9EmNT73UolFp4i8cI6w3UL5XOGgWPuwFPjOeKyNyHhxYj1GArMVmRY48jq0Y5Ui7xx7ezFhLai2//Bt/jV/+tc+jPEUsZRHOqcDTxRqtEUoSyhClXcNMoxEajDYFxuKwGcZWLiPWt1Prjhtm6vCBMFZUYiCQIQZIsxwRG4zN8TxLYD08oF6qInNNe32ThVhjQkuGJvRzLMrhlzhb0txoKmHE1OT0jzXv3h+YWgAkSjmGh8ChycZaAt8vfJEaLJ88ycTUjJOxFx6Vzg/VUS2lkM7z0LjuoRe6To1SPrk2xUl1zBGTp3hCMuz1WM9uUG+2iIqAkSgMUFbjedAbdNnZ2SKIIsqB63xiQRTeEFEBxHieA4Fu3LjJ+voax48do16vMRj22NhYx/d9mo0G1loGw5iGdkVBHMfOrzJJ8P2AXq+PsZa9dpugkMdXK1XiZOA6p1jKfsQgTlBBiUZzit7eFpuxZrreYKZUplGvs7K8SKd9L9dvrHLxyjVu3Vpzch5txwxFrEAhnE1BkU2AtqAspjhfI29NPEOlUmd/bxdrDJVyGWEteZoghCAdDPDcyUf5AcI6SZu1kJVKmDzHU5JyuYQ2lr1On2atxMc//Tms1lx+8zXMcIDQBmlHtxKHooze2mjCv19/mb/ooZTC9/1xd3NnZwcpJTdv3iQIfNf1C/yxZ9KouNJaj6VQ7XabSqXC+voGi/NHeOvNN6hPuIAp5QdsbGyQGsPeXtvdUqWk1x3y7WeeZW52iYlSmSzLsUagpEeaDxgOY6IopNfr4fslpHJdEcFBqroQBTggQNhDC0ghjdcGpPRQUpBrjbGWqBRxZOmI6/qHvus4FfNCSsckSbOYQb/H1vYWvh9QKlcB6wIyCoP0oEjYO7I4z4U3XqHfaXPfvffw9uXrvH35ClK5RVYULFAr3OuAY0EYrcl14YNqLb6vCPwAIfICKHGetbI4ViUluc5QhTWBthZpDAbjigtjMeNgNoVUo2IMjM4LpjbYwqNSSkmWZ2hrGAwHgOD48jJxkjE5PcPa6i2EUniej1I+kNLt9vG8gDiO2d3d/Uudp3/Z4y8C9DhcFB8Gv94tGRx97UL1rMkZtvcpFdelxoDvVg9f5xw/dpy9rS2uXr1C2umTD1PU3DznH3uI115/nWW5zIvff55HPvBBJhfn+M43vsn03DRBs0o2jGk0moQzAVpCkrk5M4wLj2opOH78GP/u3/0uv/yJT7G0tADCsLu/x5HjRzl/zz30un3wPGKrqVdKCASvPfcCN965yL3LyxxZWGC7vUdtssn5Rx9h+expKs2mA0dG69vP2ObozvF+2aXv97k/ymPuNiQWlEBjsAie//73+M//d/8Zg+GQwPPIJSAlAZJcCXb32/RefRXfCEyeEAXOAiUzmrWsw3/2P/w/sEbwN//uX+OXHvx1Lr36O1Rr2yhrMTakUi/x1T8Z8OaVM5w8c4Qw9Fm7scXx44sI6aHJEcoA6oczk6VFyxxlPMLcI1UAKf3+gKx5HLt8kjffvMSVjdfp1CqU545Qn15koVXhkXvuIR70ePHCO+RpjhaKSMDWjR3++NYOVhmsqWKsxwMPnOTt3V2GosTTv/9NvvTXm3z/239Ekmb829UtHj17nolSlcyz5EIRaFEEUP3gA7iTofYTA6xaVTCFKO4RBnAgTXt3jW9/8/eweY9I+nR21pibrrOzl3Jro8Mg02RZis0ThEkIyxM89vBjrCyvOFm/kkjl2HhK+gjhgZAYqZ2tByN53WgT4MAwYS2IDApWKriAJMY1VBF+4OjW7ucmRxhdzD3rGqhGYyzIoOxeSwp8KZAScj3EYHjnnSsoKXnhued48Nw9PPLwgzQmGvilgCzLeeH5l5mbP8LU3DJLK/dR8pwfv84N2tw1Ns3V8NJgNQXoCUZbdO6S5dM0Y9Afst/psL3VYWe3gxGWT33m8zz1oS8wNXWCLLdYk6J1RhIPx6DTgWzvBzO5R7LtbrdDGIYFYOY2UcZqTp88y9TU5LsksX8Vdehf5AjDgGq1yv7+/rhJM/ozZmWOWL0WJiYmOHXq1Pj83gkm/rDxvhtnxcOUdOyh2ekZsrH1hCaJ46IWFeNGJtxuA1ZUFI6JBAdMKBjbFCjPw4oxRdXtIQu/vLFnavFrRr/izjHak2itC0l/SrlcplwuO4A3y8Z7hsMy7xEgmKYpnU6Hzc1NtNZUq1VOnz7N/Pw8c3NzTExMjPcTP2/jMGj9bk/l2+tG9+d2u8DDY/R93/cphWU6us2+KtPuS7Yu3GKqJJmoh8w1yzRaJVRtgOh00MM2nvSxmUBEAWDwA1AyIB1WMFqR2hRrNadOnKC9v88NebWwICgAc63RXo42PhaXCWGMJs+D2+5xYRgUgKrztxai8KsWij/+469z9uxZqtUSQjof6SzNMVKMWbpa23HIltYpnZ0Bx48cI4oirOeChEzRH5haPsL+1g6ZhP6gR6lcdirWn/I17d/3mJmZJZyssyszQiC0ECRQK0XoQUp+5Trd775I+JknGKiQKA3AyzElS5ZrTKDwJPhhjZUHH6I5O8fqq2/TWd1C+xK/VUcpRaQVmc3RwOLxY3hG0tMZ1mYEtYjct8w9cA/BMOHCpXcIF5eQuWLv1irlWCMGhpbnEYzAfV/glwL6NsH02tib2wS9KaZnjxDXNHFjkkSVYXGJSqs13ofdbS68ax0frbGjf3JniXoIfbVFw+xQ48NYi1YCIxSownrTUpAqXSCL9BQqy1FWuMfbDAqLCzfvQ2Rh26ekZNysMrc3WcYEyxEeaEdWRo5wJox01oAIrE3QsasFtLGkWYZVHgtHT/DW7Dx5GCGVjzACbQ25MHhSuroNlyUihWBiskVrovljzbv3H0A1YkUWncARcGiFxFeK48vLzC0u4Ych+aHuoDtlByazI+aixdJsNnjg/vt568Il+sPEdf+0Ljo5FmscYzPWOVmWws4OUbnCYDDg9KkSQaDY3d2iNT2JlyboqEw5iFCeh8411mr3fowlSzM6/R67u7vMzcxgjGUwGIw7u6VSiW6nSyUqkeXOu2pubg7f94tFMaNcqbK5vUOpXKISOqPsnZ0drIXr164zNzfvwGEhqVSrdHt92rttyqUy+zt9hmnG6dkWVWUJQ0UpKjM5OcU999xDkuZ0ez0uX7nMOxcvsrOzS5JkrrbGdRtcEEHhX2pdP0FIRRAEVCplevttBFCrVrA2d+fA2DF71KSiCLHyyZOYICq8ZPKMPE9pt/cQQhAPncRLG83cTIuPfeqzCAxX3nqTpN9D4ABVDFhtCwbBwec7kkL88B3lX+wQQlCtVsf/np2dJc8zsJYsS8nydGwsf/36daanJ8myjCzLCs8tTRAEDAYD0jRje3ub6ZkZcpuTpJnbeGcZKgiL9FgXTuH7Ea+98Rb33XuBRx9poPNiE4XADxQ7OzvMzS/y+utvEUUl/MC7rVNZMNtvS613hH0HcmZ5RrvdYXZ21jFSrMUrzrkQgjAIXRhFrtE6Jwx8sDlJltLvdxgMByjlEZUrqCB07E9hUX6A7/lgDZ6AY8eOsLN2nYsX3uHeBx/m8Uce4satW2RGk1uXHOx5Aq1H1HxdsJKF8zbGFsWFJk6cmbyUbtM3KoQdI1fhqcIzqzgHUinXFSs6a1YcAm6LolkbR/FXyjVrlO+hrCgaPAIzOubBEN8rs7a2zvTMNNOzM0ihuHnrFiCoViqO2WoMg0FMqVT5S5mfP6tjBFyMpJkjlrpOhuxubUGakVmDcYY9oBRpnPLWy69x/t572B/2uWfuPoIjZb72lT/kwjvv4JUjWlMtEmuZnZji2af/jI9+6Qs8/MSjeBYCK0H5aF8wSFPe+P7LPPTEY44ZKywUcjI/8FleXsZaQ7PZpN/vce7cOZ5/4QVOnD7B62+8zqOPPY4WEhWFrkkQp5yYP8KRuTk2NzepL8xw/wee4Pjp01SaDYTyQKqxRc3P+rhTsgjvZivdbfw4csb38xwjHLtP4vyP/9n/55+SZyle4JNZWzRtNJnWWAllSuS9LgPjmHdeWObI/AJWa4bDfbQA3bL86z/8Kh9+8iGqi4vowT5+rkmkz0uXBjz/WkYvq1KdmWbYTXnzhTdYWj6CCSTKOMm2tIfbi+9xfAg0Ei1zIq3JRYlURlx46wIml+wGU6yqPaorpzhz/wOUbu5y6Y3n+MabzzHs9kh0jqk1idIcL8kxRjHs9/ErZcJygEn3sTZga3eb737/u9x78jTffvpZ/qvf/MccOxKhPMX+/oBrN2/SPHsKhJM/GelYkOJ9zOyfROafqzWd9JLRPdZKjM555tt/xLUrr9KqGIx2PvVS+UzOVWksLvDymxdZvbqLiTWVwOf86VOsLJ6kpMogXGiBexEHKFohEEVSrVsPPLDqEKGjYD8Vz3EbcBfK4pieolBsFWGBY2961zzNtcGQFw1zhRRlIr+MlQOktAy6XeIsZX31JoNBh43NdaIoYuX0WT728V/CQ/LCiy+wlB8hzTJOnT3HiVM+fhC5AE1t0DoHbdEFUMqhTdgo5Ml5ubuWRSEIZCTjTVOn2Npv77O7t8utjQ1Q8Ou/9ht84IOfI4zmSFIXuprnlng4xPOc1+cIfLk7iDoCXFydkySJs4Yp3qYLmeozPTPJI488MrbmGv18dBg/KfPyRx1SKmZmZtjf3y+URQfg4MgzVUqJpzxK5RIPPfSQC+rl3aDL3YDSH/bz9/qZLdZXYBx4KSzYXJMMBmRZjpQC4SmyLDu0Lxip7QoA+PC2/hARFwFTU1MH7FUpXD5FEXx6Zyji+H3dse6ONuFpmo7Vh61WC2vdnMrzfOyJeTuIasiylE5nn7W1NYwx1Go1pqenmZ2dHTNRR16sP+3z7Mcd754740/yrnPnvZ4vhAtOKwUBYVihG1UJHn6KbH6b7W5MOx5ydbePPzA89LHHSNklFxVM2MTXApOkEEhkGKG1U6vmaYIhZW9vj16vx/T0FJ6vSFPt5pzR6GKfYYuwP7e+abTRxapH8f5c2KuSXnF9OSnH7m6X577/KidPnGZyquXqAEQxryx5bgoWv7OJCEOXF7GzvcOplZMoKRnEibPBsC7MO6hFhLKFwe1RxXBIVCq9h5JnTGf6kT6/n6VRrtY4c/99rP/Z0wS5QRgwRpAa8EshDHoMXn6d5uIk1XtOkwWqCASVZL7EaIHSghSDUjA/P89kVGF97RYbmxt0dtvYTENUJpA+SiqsLzGZIcBDauvS5q0gqFTwSlWOVR4ijWPSrX3iRFPxA3bXNslTTdVYAiWIsBhPEpdCqtJDDA1605KkCaVWyL6fIhbreMerBJVwHKh3WAl8eC26k5l/12Fh1BgYLbzv9XBhBcLzMAZyiztu40L6DA6oBPd+0qKp5tZXjygMEEqS566p5thhtshaEk7Frrxx8Noog8ka10RW0mUESWtdphBOGaSNZZikaCPRVpNqS6gklCssf+iD6EyTlytOQJhn5LkmlwLpqwJgPghqHClhftTxPpmpjhF3GPABJ1E2CBaWljhx6jSVeuOALSYcGAQHbCWLKdhzubupNep89jO/jPI9vvmtZ7hx8yZJ7Ex5wQFXnpJo7ST7VmSApVIus7e7zdDAbrdLrd6g0Zp0HgoNyd7OLhPNOjY3kKYEURnPl+zt7SFwxulZOmR7axtrNZ7yCILALbb9PkmaUqlW2dnZIc8T0jTF933SLEMbQxhG5Dojz3M8zz13YXGRJEnYbe/RqjfI4oQwCkjimMxYBqnhjUs30IM+9xyfxQ8EXjBi/AZOkqMkDz/yIN1en/X1DS5evMRbFy5y/eZN8szJY6xwtGmjc8KoRKvZIkkS2ns7hEFAvV7HGkOSZuOiXmcpmYUQC6QI5aGCkCzLCaKS65Llmn5/SBJvoHNbGHpb8vUtjky3eODhR2lVK7z83PdIBgM3B+y7b5bjmyI/nN3yFz3iOGY4HNJsNkmShFaryfb2FoN+n+np6XES3ESrRRiFDId9hBasrq5SrVaZmJigUqkwHA7xgwApJFrn+KGP6Q/ItKOhC+XR7w8Jg4A0TfGCgEG3z7ee+TZLRxdp1qeKAk0VMiRI4piZmWnyPCu6ngcbCbfYWeR440UBVhZbMpszTBLXTbGgfMeu9cNwvMHLM81gECOEJPQ9siym3+tijLN2KFereGEIUiGsdeFuQcAoCMOrVDl27Dg2S3nm299g49YNji8scnblCK+8fdklIipNXnSRjLbOS1kIbLGKKKmKLlSOlMqdmyKRchTQo5Tn5rxUY7bAqDHjK98x2hFI3xUMWeaueRBjM3ghHEhtZSHzF44FWY5KWOPYRdgyzVaTa9dvMj83w9buFkIqlOcziFMW5ufYWN9gr73/C4P3HzLeDYodKtzG9x/3hSzmchoP6a6tO7axkgzzhNAPqfuRYzF7PidXVkAKFmfm+Df/42/zpb/7H3L2wXtZPHKU7k6br//+l/H9iHs/8DjlK5cRac4L3/ou5x+8j7BRdXMjdMVMtVplr92mXK46Gcuh9/zAAw9y9cpVatUK7fYeW9u7bi4hqNcbSM9HhQFWSESmefvlV1manWF9e5P5E8vc94HHOXLmDKV6HUbXtBidiQMbhZ/V8cOYqe8Fmr4Xe/FHYase/h2jrjlCIoXzPPoX/+K3eOfiJdKCaVRvNBFSkiYpQaC45557mDi6wG6nzbe++jRl5SN9yW57j7/98c/x0PQSR4+cYTXaZfnhFiW9xtrqEOE32ereZLPt8dv/Zotb7RKtDy6w8qkP4tXqzMwvgAjxc41CYUWAQf/QPY2wAmUkqa/JMKAhUGX+ze9/mSTt0d4boM/cR3nlJHONJr/3X/yXrL7+Mu2S5PjZB6msLLOe7cPlq+jL16nPTCIGIR/9yOeoz1reeP73eefCHpfX1rHDPoOba0wsTbO5s8axlSNcuPw2vizx+3/wZU6d/odEKkRIxyYacxR+yOf0k8uQsYxqT2d+Y7l66XXefPVZdrc32Fxrs7AwSXNikZuX12lOVllYjvjCpz+KH7Z4/sVX6fVSjp9YxvqSXLrQRIN1oYsKx0CWOVJ4TraPKBQlYrw62hHoikXggSkSwA2FzF9jTY4lw5A5Ox2rMFaQaYvOJYgSnvQohT6etAz7u3T2rrC/v8PbFy6wdPQ4nU7Mo48+xblzD/PWa68hYsMLLz/L+cce4/EnP0yjMQk4myArLbnJnJ+pzbFGYuyhaq04hsNsRsd8VIAuQF9T+A8adJ7S7/fY3NpkdW2TqOLxq7/2t7n3vo/g+RPEaYK2Gq1NEcji3cYqfK8xYh8qJel2OwVwaDE2J04ca+zsuTMsLMyP5ex3Y2H+5M7R9zdGBIGlpSWyLENrPbZ9Otzk8gKfY8eO0Wq1irCOg830qGa62/r7wyT9h79/2ya9mOVj4DHPiYfDon4r2Ngj69OCOSuEYyyNduuj+6YQriHO6HEcAPa28NazxnLr5k3CwoptdDyH7wemYHHfOZIkwRjD1NQUSiniOC7qyoMws5FP5simYGtri42NDbIsodVqMTU1xdzcHIuLizSbzZ9bJup7jTsBnDvvf4cZq3d7Hhww0pRSeIGiUg1QwPyZU3RmZmk0Jqg3GwhrKZdCOiYnrFSIhEe3u08cKHa31plqNAlKESbPUTYnTWOiconp6Wl2dnaKfX1evIHRPHT7mjzPxn97foDva8cqhfFaXrElosgv5rWbcxfeuczVK6v0ugkLRyZo7+3i+YWfqrAEoY/nuR2eUo5l3263WVtbo9fvI5VHrz/AAPVmAyV8EiFQ1RKR9cbNJAuUy+X3OIe/mI8AnvFYPL7sAD/tALMcSZxoqpUS0ksY7Gyx9/VnaUqw950ufMhLaBVSThWesWQljfUNuRF4rRrT06eY7h8hvbXNlStXWdvdxU8HRJ5HHkLJ9wilQGpLyfNIhfM3l76PQlDyy3ilBv7MBFpozNIU/Z0O/Y1NKmnCBIKq9aikCm0lWSWirQL6jUmEktQXZlk4e5RqJUCpAwUwHDSNDg83Pw8a+3fW4aOM3BEl5DDD/87fI0f0rty4ffrIV7hgllqjGcQDrHBNYKfkcUQ/gQIBuc7HzHOLw7OcT7XFaPce/DBwoWuBzyhnSXEQ4ufwAucM6wFhGJKVoT+I0caiPAeqal8xee89CK0ZFmHVOs3IjSbzJFLkWDWKXbVUqxWM0T/evHs/D/I9bxyoIGXBM7UCI9xm9OSpM7Qmp1x3HidZtsZgdGEJbq2jMQvhkr/zjObUBJGn8Bs1fu0Ln+MjH3qKa1ev8Y1vfJ2337nA+sYOUlAAMsqh30KQZxk721t4SlFqTDDY77B5a5UkTpmemeXi5iZT01PsdPbd+5YSbQWh9kiSGN/36HU79Lo9ssylSzbrDYegK0WcJihPsbq6RpqmbmNWr7O9vUO92aA/6FOr1dja3mJpaYE0yTC6T5wkDOPhGLWvRBEV30O2avSHKWmcUKtNsLbXJ/B3WFmcIfQFnhBEkWMICqnA86jUqiwuLfLoIw8zSFO2trd5++0LvPnW29xaXWNjcxupFPVGlXZ7G60N9VoNKQSDbsexbbVGKlkYqTtGsc5St/UfXQhFMqjveQz6gjAqMRz0yRNNvdUkjof0OjEm7nHfyRU++uRjBMLw+isv0x8Mi5vjQdfuoLjBscXNXa7Mf49j5Fu4vr7OwsICa2trjJJvrbUMh0MH4BWesdVajW63y8rJE9gi1ODGjRu0Wi0GwwG1uQY2NQyHQ7BOcuUpSTdOSNLMhTcZ558TRCE3bt3ixZde5GMf+iV86RVsbhcuNegPqZTLtDtdqtXqbb48lqL4xLE2R35QFB3zSjXgeKXqTmohwXQeqsbNGev8R7vdLlEYkQYeVqeUKyX6/R5BGOIFDgS2QhB4vrO9wAWlWCXQeFQaE5w8dy/K83jp+e/RrNd49N5zXLu1ST8eFtc/gJM7CuuK+tzmY0nPSCrlppdjrqZpOmb9er6HNbYAXIv0WSnReXZA5de6MK224+4bMPZgGrMShHIG8tLD4t7Lffedxw981jfXKVUaVKpVNra2mJ6eZjCI2d/fxxjD9u4um9vbWGMoV6r8Yvw5RrHhvr12KzZh1mJ0zvbmBrUoRJgcEadgNH5ZkWN5+8olThxb5uUXXuTe++/n/vvuoz7RQCrFtTfe5sIbb/PEJz7OBz7xMTZ29rl66QprN24ws7TE2dOn2by5imiUaXklpBXE0rCyvIwp7hNWFGwqipRtTzC/sMCfPf01nnjicbTRLAUB2zvbREFEmqZUGw3n453lxL0+ejBg/vQKDz75BEdOnSSo1VDSKwzO3VozIiD+ooT9weMvYtN5uAgcMZyUgSTL+M//T/8X/sVv/UtnOxQGTDbrDIdDRGEpgjG89fZbDC6+Sa3V5MmnPsB9997Hv/zdf013mPGdb32HX/q7/4hmq8XbF1/j5eRpZubbBAaszRF2gd/7ykW21Cwnf/0jJHWfkq1iyKnPNAkSixcJMunsh6SLFvyBx2ML03+lA4QxeNZi85RwokbF9Ln+4hssf/bX+VStwdU//SO2bl7m5Jlz3PPZj3HsoXv59gvPM+9PcWVjn7yRE9WrpHnO8vmzTC1tI3ZarL7V42O/+mtsbm7yxiuv8aUvfoJmPqTT26EaVWjVpmmniqf/9E/5zOc+haZMKU/RP9UgQZHsioVCF5WlA1554Vvsbd+itz/AGMWVS23ifclks0yjOkO8M2B37TsMMssTj32cfmwhCJA1HwKPTANWIEWAVD5IiVIuSKFwagRcjStGaiJrGLkoW6vBpAgSsAnGZK4+0hKTW/Lcoq1rTAoLvvQIA+XWG2nY2LhCmu7z2qvPMTc9R6s5wSMPf4DcWCKvw4vPfZtqvcXU4jyzMwssnrwHz4twYbzOWsDqHJkbpHHMDm3V+JxBAZxKizC3X28IgbASiXUep0aD1WBduFAcJ64+bjT565/9AidPPoI2LYZxjiFz/rNxRhQEgCXL9LtAlcP/Hu0zrLX0et0C+HBkjDzPmJiYYGpqkoWFhbtazoy+/mkHUkdDSsn0zAy+77O6usrOzs44GVxKp9Kp1GocPXqM98tIvfN8H/76/TBY7Qi41E4GnyTOTswrlHK578CqTFgOuVYVPte3v7YxxikejbunCuvqiINgSwe8T09PO1uBu7y/O5mqozoxz3Oq1SpRFDEcDp2VWu4IBo7kM0qYdzXpzs4O6+sbZFlKs9liauooi4uLzM3N0Ww2xzZi7/dc/ayNO4kz75b4O/766Lze+by7NVjvBgIFXkjklUkTg5ABMighw5AMQVCOSHyFRaGtj01yVLXO0EKlXiP0A3wE0veRUej8IYV0qkLlfFOV9IC8IAQdJgmIccPCHPLeFQikddZC0hokAX6gUNL9/E/++E8Z9GOuXbvF6bNH2dnaxJiMwt0CFzBYqHKtQSnJME7Y3dtja3ubLCuYeVh0luNJRT9PGQ76YDSNhqtpev0+g8GAVqs1bgT8rKxzf1FD5uBVaoiwhMlSMinJrKFqJDbVlEohuU3obayz9Z1nmPBzKufPsSctvvVROeSes5rz8BhYW2CSHlFUpXa8Sm1+gYXdNns3N9i8dZN4uE+SOrCw7AeU/BClPMJSGYnCegHGGjwLXikglRrbbODX+pSbkyRra/TTlEQIEhR9K8hrDez0JNGxeRbOn6Y5P4tfKhHKcBzIfPh6Gv093kdbixRF6j3vcb3ZAm8Vwu3rxMHaebhhZYxx0npjyeMMK2CYZWRGU/jwEfg+0vfw/AAhnILWWkuSZggpkNZluVDcO7TW+L6zAFQqGL+eVBJtLYHvFZy9QglvLcJTRQPPoo2z74x8yVBJhrlBCI3RhgyB9XxUYfupDOiCNZtkmbOqtF7RuTMcPXoE7O33kD/veH8y/0Ja61hgFikUBoHyfI4eW2ZmYQmUcgmjhY2n1hqbO5q8MRoK78c8TVFS0GzWnfGtdRTniUaN2vnTnD65zOraGpeu3OAbX/8616/foN3ZR0u3mFkpXVouEADK5OysrZIMByTDAV7gFjmARqOBzjUNC/t7AzY21p0HKcLJnqOIYTzk+o3rTE9Nk2cZg+GQiYkJdtp7lMOI4WCIEJLBMCYolQrPHk2SxONOsS0218NhTFQqs723iz81ze7WJiJQ5Fq47kWuGcQ5F29uUas2mG1FqECBdcbW0vMQBajqKeXsAiplZianuOfMWbLPfZbt7V0uXLjICy+9yDvvXKAaOJ9KnSZk2pDprDBqlxhryI0BUQBWOCm48gNG3qk5loHV1CQkOiNLc7I0IwgkXhYwHHYx/T3mPvAox44u8tGPfoTJZp3vff855x9bAFwjBsNIooX9sefmn3sEUYSnPGZmZtjc3KRSKeMpSRiGdLtdABYWFojjmMFgwMREk3K5TLfbJY5jpienKJfLVCoVdtu7VKsVrl3eRAaSMAqJgpA0jl2AUa+HHWnehJP6WWv47nef5dTKKeZnjwAglY/yLIP+gGarxdbODnEcE5UqjDyYXVeeQg4v8JTPzZs3mZmddeCjtQX7BUasG+V7ZEk6IgePgeDFxQWiyKNaiRxLcDDA90M3r5Ry4JKSReHrigSDxQiF9ULCiuL0+fuYmGjx9msvMzs9xbmTJ/j+Ky+DlHi+8/Wxxmmy8jxDePK24gjcIj+Soo2+N/ZRpiAhCJeQaQrQdHxDoEgbLOwB/DDAaI3JTOHJ6hXymxHJoWAhGc3RI0fI84xet4fySpRKJbTOiGPX2R11y8rlMlJJ4ix1C+4vxnuOHyYTGf1YOtkBw16XainEM4LhMGVrdZXFI0t4nsf6rVWksXhC8Njjj/L8a69Sn25x9fV3aNQmeOKjH+aZZ5+FNGNjZ5vjp85S8nwuvfI63f02tflpyo0aaT/l0uZNjj9wDxLpmKa+h86LuSAZJ8eOWE6PPPoIlUqVMAzY3dulVC5TrlTwgxBRNFlu3riBxbJ44hhnP/goSydOENWqbm2+g4P68xKs+sM2jO+3kP/zFPx33ewXXTpHfHFf/6v/6bf59neeJSyXSHKNkYphnpMZgzQuzE4DVudUgpC//uTHmT26xIurV3jsYx/hzWdeIGxN03rsNMPMIlWGp9pYuU8uLIKQN15Nud6f4vTnPkCUlvmN+ceY+N46zdkjVE4toXVMJj0CbRHKvd4PF2VYhDBgo0LFnXL11g3i3DCzsIgUgmiqxc1Bl6PLJ/gvf/0fUt0dsnrPOXrTDS5+5RvsrW+SJwm1sESzH+JbhQkTrq++w/r1LT752Cf4lY98liuXrvArD36AJBSkZcnFFwf0e116SyWev2J5+Zuv8cuf+TSesRhviDBl7vKR/3QMAa74KBgX1rKzeYtrl1+n095hGBtCLyo8wGrU5le4vt9johHSqE0yoSTXXvkunhfSVREV39JoTFOqTSCQkOd09/o0pybd/SyqkREhRQAIJI71IMkL8NSANVgj0doxPHUuMMZ5h+lcu2f5HhKDLCSH/b09hI556flnmJ6bpN1pc/LMWR5/6sPkA8POzh6Xr17mzPnzhPWIR84/ShCWkMrHFuoOSwbmgNmHleTWNX9HFkJSCNcUwj3ubpJld1SAPWgRjBoaQgZkmWDl5L08+uRTLC6eJ8tCcpOR55Ysl+R5QuAHgCLLY4Q4kPnD3VmQo7+Hw6HbgAkny56bm2VlZYUoOkipf6915acdYBgplizOYqI1MUG1VqPf79Pp7NPpdEnimDhJmJtfIIpKt53TH0UB8MOec7jO1wWQmqapYxMqRVgKD0JphQDpZtoIbBuNw5/xiMkoxixB97dwHfqCze2CgTjUTO92u1jrwoHuBPWclDpkYmKCNE3pdrukaXqwMS/2LaPj2N/f59q1a8RxzNTUFLOzLlhqenqaVqs1lvPfeY5+2ufYjzru/PxGIOEISB0DksWfO5935xjNKyndniCJErQwJDpn/dYGr3//JfwgolqtMNdo0arXqLbqVOZa1EoVB7qaiHq1SqAkuc4ZZk4hF5QipAWlPFZWVkjTlN29XXq9ThEGdRhQLd4nIzJIeuhYDUKMQgM9qlQQQnLj+i3eevstjIWn//QbnDt3kpnpeXr9zm3HFgRuzRLS1atzcwsEYUSlVqNar1EKIlbX19F5jpUKhPOAjIqsByGcstIYg+/71Ov1267Hn9e5eOfQQlKbmCQqN/C6bTSQeY4x6mcpKA+/XsLrxeTXb9D9sxwZZzTO3UNaVsTlOqln8SxkucHzFDK1+Jkgk5D7CsoBrfIck4vzzLWPMtjbYePWKu2dXYb9Id3uviNSKY8giFCej/bAN5bQuCyfRCl0xZD6IUmg2Bt00IEi8wNKjRYTs4tMLM3RWpgiLIUEpTJGRFgZIUV2G5FtNG5bm3AS/Dul/8DYg9wYgxT+OBTtMBFqpCCx1pLlGSbXTp6vC7u9UoD0I7wwRIHLO8GFSVvtrikhBL5SWOFec1iQsoIwcLgHjG1gxlhA4I+PwNUtDvF1VgBqzAhXxVISeh6hVGQIpDWF6EdgJA77AqQwKM9z1knWNc4CT2G1oVapMDM1yY+p8n9/YKoQOEbqyDxaSqRStKZnWTx6nLBUJckztNFIhNscGDv2IMlzjRdIlOe7r30f3w+LRVaAsUgMSjmAZ352mqmpaR59+AFWV1d55rvf5Zvf/DY7e/uFR6dkY32VRqOBLyDud+jlKToZ4EUlAiUIooiNeEgYRGAhHvZp1OvoLKPT7qCkk+L5ns904c3T7/cJwpD+YEC73cabmOTWzVVm5+bY2tmm2qihPEGaxpTCkN5+h9D3SFIncXEWACH9ZMhwOCDt9/BLIVJ6hF6EEFCuVNnf3eHSzU1q5XmUlHhSo7R2gRieQgoHQqmCWi2LEIUgCKgtLrB8ZImPfeiDDIdDtra3uXjxIldu3GBra4vV1VV2d3fJshwhFVmeOmlqwV7MMo3OM7JkSBSVQCq8OCAb9LBWoDyFkoqOSV2XLM/waxUalRIYw5GlJSaadSYmJ3n2u8+yuraGNtoVyHd0H9/L1+jf10iTHBEqcm2Ioojp6Rlu3byBtYbJyUlu3bpFo9lkd3eXSrnM+vo65XKZPM9ZWFhgOBhw5IgDQRcWFnjzzbfwC5n+cDhEhyVKUUiea2cF4AegNUmakmYJUeDT7uzz7Pe+yxc/P1skcymU9AjDkL29PVrNJoNBn2qt4Vh01ple+74qFgp3/mbn513QghktSu4YrRXF4mJRnirAI8fDabfbHDu2RBgFSC8gTWIQCitd+BJC4imJlAJPOtYsIwaPsFirsEKgohrzR09QrlS5fPECD9x7jjfeeYvYGlJjnFeKdl1/zx+lTxrHohHCWSMU17czhjYFo12MJWpSuAV7FGrnK49cazTaAf0CpHINmCROgKLjVhRvUkiMAJ3lRZFiadRrGKPpdbpj5m+aJCSDofNP8n2CMCTXOf3hgOZEi85+Bz9879TbX4yDcQBqFXL+8aR0c1AKiIcDhr0eke/T3msDsFiE/elcMz+7wMZLL7C7tUW73ebs6VP4yufUmbO8/uorPPjko5w8fYqS51Er1fj9f/Wv+cJv/DqPfeSD9AZ9Nrrb1FstpqdmWL1xg7jXo1StogKfmxtrDHpdJqcmUEUTIgzLeL7i7beusLK8zPb2FnmSsrm7w7GVZaq1OjKKHLfMwuXLl6nPTnL2kYdYOLFCVKshlGOkOu74IZDvcDzmz9H4YXK99/P9H/b7Dz/XeayLsXwV3Ab4X/3rf8t/+0/+39y8cYtqvc6JE0epNOvMLy7w4gvP097Yoru/jyclXuDjScl3v/MdHsoe5fjiHK9fvMh/9Hf/I26+fZnL1y+xtHSeo0ePsb7xbcQc+AzodRR/8No6pz79t1lvr/G/+OjnuPdPLjJY3WG3mxNJEMmA/dhypBKSG7CjxvMPGMKOmsOaXCkUlt3OPrsqoNxocd+RRVYjn28heHhqkXLDcnrPMHO5R21ulvvve4CvvPJbUBK0bYfkRpu63+B3/82/pdu+wHC1zdl7tjlz4xJpnnL2xClUUOab2+vUz91Dy54hOrHE43OXufKnf8C1y5c5vXKeHA/1A9/57Z/te22O/6rGoauzuCxz3n77eZJhh0FvSJYLomrAidPH+cSnPsXkzBxBAOmwh+6nXL/0DmGjhMkTjtWqdG5e5q3nvsuxo8e5cu0a5+65l9ffepOHHn+SXpxw9Pg5PK/O1m6bZmuSeBhTq1XIbU7ge+Qmd55fWY6whiQZkKUxWeHTP+ju02jUuHrtMsZqXnnlJZ545BEuXbjAfffew2NPPMbWzh5DLdnc7JDmewR+nWMr57nnwSdAOVsAIdxm21q30cHkSAzCuPpFW9CFmqyoKBBWjjm8P2yM4VfhWp3GghEeyDKLx85z5sz91BqzZJkizTQWi84N6dAQRBWESEnTIYIQKxxrcQRsjT+xQ8CAMYZut1MoV9wasLi4yMrKMgdLyrtBkJ+9UUiJC6ZQEAb4gU+z1Rxvhp2CbhT0cSBVPAxQ3Y2p+l7s1buBhrdbUt3OhiuVSgghiKJo/J5sQQ5wG1TXFDeY2z7f0WuMJPqHgXprnALLGusYB2YErB6ci1GolS4UdofBu1YR0NLv90mSZPy+XH2qC1aqA+uvXLnC3t4e09PTnD59mqWlJebn5ymXy3ie9yPdw36exmGG6u0s1Xf/uXMcnmu3zVdpkX7OsLPPUqVJ+9YasTUEpZA9GVC3TsG2V/GpV6o05yb55G/8ClZatJDgBUReiJSKJEvpddokSUKz2aRWqzM5MYnOc+JkSJ4dMFRHAJWbH2LMbnZS5AJ4tcU+DEPJhDz99NfY2toEYH19k9/8zX/C3/gbn+eJxx9mMOjTH3SxGLKsX+RVuNyJZrPJsaNHSbWhN+jjS4/WxARRGFLyQ4ZC02g2qfgBxloq5SpBGBXXjst8CcNwbFP2C0DVjSwQeKUyUbmG5w2wZGQexMLiWUNmDGV8mpUyg3ZKdmOdbvoczb2U8iMP0AktBg8tPayvEUbiGeFs/yzkUoMnyIRTUITTdSpTDWZOLDPs9tlZ26C9uU1nZ5d+u0u7s+8IalIRKkUuJZkvGIQSpIf1IZqbIooWiSZqlJsNaq1pGvUmXhTgVTwk2lmeeI4Y4Kww7W3XDdzRoBQCKb0CZ7jdVvC2Zkeej4lJea7H8z1NU6QU5IWK1PcjKtUq5VIZA2hpyK1xdYF1gWvWuOtkFMQphCxyhwyZyRHS7R39gsA0Cv+WRZChC5i2BbA6wh4dqKsLGb5UrtEgij0ZnqvvReyC3qy0aANC+C68UBQN4YJUmGUZxhS1hzU0J6aoVysFzvajj/cFpo4kN6M0bgNUKlWWT5ykPjFJZsCMuqjWorV1NFopHTinFJ4foIIAawXNZsPJk3EotpNYuAJMSoGnJLlJwWbMz03zpV/9Ih/50FP80//vP+Otdy4yjFNk6ADVShjS2Rti0aS9nDyLufTWPrXmBMoLqNUarK+vI7FMTTTo7rfRqS6SxTxq9RrDeEBUKjE1NcXm9hbz8wtEYchwGNNoNAgCnzCK6PX6bG9tMDM1hac8kuI8pEmM73mUSyVnsK8N6TBmY3uH2ZlZSpXI+TNYQykMCOfmGKZD1rd2KR9ZJM9NkYyqCX2BUAohDtgAFDyDEWdPCkEp8glDj1arzskTx8dJZmma0ut22dvf58rVq7z+xpvs73fY3tlhY2MLYXKMduEBNk8RUhWhYiA933lcKA8v6YMQKAHnH76XE8eO0NnZIs8ywlKJRx55hKnJSb75zW/yxptvkqYJh/vKd/Mu+vc9+v0+ExMT5HlOuVxhf38fz/OpVFzHfmJigt3dXdf5TFMmJifIs5xWq0We5+zu7dHr9hBCUKqWaTYb6DRDixzfDwg8x5DMsowgCOkNMySiCFQIyFLHBHnppRc5deIkDz3yAYSIUNJilCGJh4RRSJql5FlemDMrPE+QpS4Vt16vY4whjLyiW0RxRou/hRjfyJ1hspsfWZ7zxJNPEATKgaxCMkwzNAJfqjG7QUmJkiCFxaKKHGwnJcRakMoFowhBbXqJZRWRvf4S58+e5tmXX3GBZcaghEIV/mdZkc43AkeFkGM/YWcUrR2z1LqgDdcNo5CoZeOjGwVJjQBWrGPOVsolssz9PlcAD9x5kKBzTUoCyiMKQ2amp4hURjIcICxUyhWENWxv7zBIEpeaKZzXSqlUwg/8H9t4+udpCOGKCsRIZgJgkRbiwYDhoEclCsjimKBWZndji8WFBdrbu1y8cIFzD93P3NIirYkJmo0m3/mzbxA+USLxJSsrxzD9hLdffY32sUVOnTrP2Y0tkjjh+888w1Mf/wiBUrz4yst84Jc/wez0DOl+H78UkRpBtVbFmgSpLFK6dEetUxQeC4tz/NmffY0HHrifQZoyMzeHX4qIU1dgJ0KQ5pojy8e4/757mFmcw6/UXFNr5EFXKGEsIE2BK4j3QUL8KR8/TCb6XuMvQgbpni7HfmrWWt5+5x3+X//NP+bPvv515yXtK4Z5xtVr1xHX4bVXXiHOEo4uH+Pkg/cw3WzywreeIUkzVju7lN98iwdkyEy5wcbeHl/8e3+H2TYI3+BXW3BrmXRwiyTeo7/pc2PbUtvvcjaYYGpxlm7pEhfOL3DZaj5kSqz4itde+T4zT0whhRpbtvzA48IBWx451giUcGtjXGvRa07x6o01lhvHsQstuvUGr32iTnBlj+WXLpO/foFPfPYpnv7d32Gv00coj6ysiYddtl6JkTan2zOojXWevfwmtekZyPucCDzOBhFq/TK9TLG2+iY1vs+Hj6TcvPgmJ049iNIg0O8TYntvVuFf1bCCcSFvgV5/i1u33mI46IJV+KGiVPf50Ec+xMmVY/i+QsgasrmMFhlzp87T3e+RxQP+8Hf+GUuL80gvJI77HD0+S3vjIs2S4e1XvsN+PybZXWdzfRPhhRgDpVKFqekpLr5zgbPn72FqZpbVW7doNsvEnTbbe7s0ag1ef/0N7r3vAd586y2On1hmOOxz+uwZqk88QTwcMju/wNVbO+zsXufhh59gsSGZmZ/DEwahIox2GxhpDdZmCGKM0QjpY41E4CO0szzJRY5VGqFzAiuwOJ9ULXyc1FXfBlDeySB01ikukMVYF+g2iDOGqaHRmmdh6Sxe0CDJwNgEpCFLcrIsISqFSOEaxq7OPJgvh708R6ET1h6ETUkpyXWK54VMT8+yuLjg1CrF+wT4WccORtfh3ewMRp6pDljsjW2VRiD14fN7eCN9N2bl3Tz34Hbg9fCGfeSTOvbFEy7jYlz3j5utRSAlBQvaFrZWh95XgU+555pDLNaiXkSPguRuB+tqtdqYWSqEcHYHlQrVapXhcEi36ywiDge0jF4zzzPW19dZW1sjiiIeeOABVlZWmJubI4qiH5iU/fM47mSi3g08vZOFOnr83c7hnfNwNP9G87ZcruKpEkrElBsNzjxwL73hgCSLIcnY6XYxAVgfMpMW81Fj05TMZpRKEVnqQCcvdCzOtbU1nnv+eY6vLCOUJIpKSKmI5dDtX22R9eeI0gWgOgKcNCbX433JaHS7XV568WUWFuZp7w7Z3R2ws9Xnv/snv8W3vvU9PvnJj3Lu7Ek8H3KTjQOPo1IFP/ApRyE1P2QwHLKzs4NXjvB8j9Sm5D6sb26wPL+IxSn7LI4U0u122d7eYnFxiXq9Nl4XfwGouowRWS0xed8Zdva2KMUG3zpWqMSxTXuDPqFfI6w1SPKUdH2H/ewl6OxQO3eW5rEVklKZYSXAKEmuAjAGqZ1XZ26cl64UYKRPIgOkUgS1BgvVBgvHlskHMcNOl0G/T6e9h+nGzv9ba4JAEVQCwmqNoFKhVanTiiqE1TIqdCrSQPlkHqQCZAGoGyAQprjvvTtwChhfRyMpvWuwOWxtdK2OVAXWWnLj1mS35rrVOggCKtXSmJHvcpOcIj0riFDCWKRx9kFCSQdcWvBVASsW11Kv1yeqlAiUh1Jlt/4bS6/XG2cOAVgh8KULmqawBbDm4F42WiO0ccGZjtgpHCHQD1CeIs1zrNYIU9jHKEUuDbKQESql0NZlMCVpSiAFZ06fQolRxfijj/cFpo4v0OKu6Hk+C0tLzMzNI6RXZL+6D09a56UopXTfHTFZC1+FNM/xwxBtXYiNLRBtIZysQ0iBzQ0CgycFWa7RWU69XuV/9Q/+l/zBH36Vp7/2dfbaHcdw9STCGkyeonEJjCifWEosinql6rqjxqCzjND3SQ3OL1Vr4tih5kmckGQprVbLeVt4HpVylbW1DbzARynFME6o1xt0Oj2iKGT11ioTExNoY/DDiMFwwNz8Ahu3bjK3ME+aa/LMEPdjZKSRUlBTJeJkSD8ZsrmTcmxxASMMWaaRvmOoOknpqB4pYokKjzUz8kRCjD8T6Qk8JFHoIUSZmakJLPDQgw/wxc9/nmEcs7GxyaVLl3n5lVd4/fXX2e/s0x8MMLlD7YWnMDqHPMUohUwkURixsrLCFz/7GRSWPMtI8wwv8KnVKpw7f45qrYYfBDz/4oskcVrMF/eZ/mWParXK9evXWVhYYHV1FWNcuJhS0sm6paRRrzt5npRcvXqNVrNJrVaj3d5HKUW1VqNSLrtE6CSn1+2yvrbO0tIRhJAMkwGzs7O0Jlrs7W8Vx2nxlEILjfIlWRbz9NeeZvHoMjOzx1zyozLEAnSRIjwYDAjCEp6nAY9BIWcbdV9uL0osY5Po8Wl1C6bvC5LEFbUIgxc43xKjne1EEPggnMG/LLrsQrhYjky4UIpRNSvGBbAA6SGUoDkzz0rc5cbmJq+8/Q7DLHPXs3B+hHEyRHqOeu8HPjrXt3XHlHLrgJIKYSy6YLDmeeqo/haEKApiWXgyC+X8loXAaIsuivaR9EBKVTR3BFEUunA8qTh96iSVcgkyw167jZABFsecUJ7HQmuCfjygPxg4ULbo7P5VAwA/DeO2jZgoBO8C1yFEkGcJ/X6XoAgl86KIqxcuMz85wyBLkIHH/NwcAthd2yTu9Fk6tsT9jz7III8ZbLd57a3XePyzn+ahxx/jpe9/j+mZRTKhKZciTq+c5PlnnuWxj36IlXNnGW7scuXKFR586knXmACssK4AKIz+g8DHmkIi2WrxkY9+mFq1Sr/ZAt9jqDNQElH4D+d5ztLRIzQaNWToo6TvpCsFiwVZpFkX14q0P/tAKrw/hcEP2jS9+8G3YSqHQOmiGBw/zLoCzTgfsQsXLvKbv/mbvPjKK6yur4HWZAJs4OFhCHRGSUpEbpiZneJ//b/9TxlKw876Jus3btFQAZubO3zylz5JuVRhsVri//bP/ztOf+Ip4v2MCZuirWJYPovtbBFGZb792jqD9Qx/e5ezDzzK9rUNto4t8PrCBDcaNfzVffZUwr99+mucOnGGozOzYPX4/R8+nkOH79ZkMWoYQULOzvYWvXYf74mH6R2ZJAs8pr0QTwqqe10uvvwCz3zzW9z68k3y2TJhFDCvqqRGYnWP1EtR1kMKn2ZrCtnOef4PvkYWhfxZPeChR8/xH3/uP6CyPcWlC5c59cBDbE5M8eLTv8tK4hPkAkv+I/fnfxLWUVtshN1bMey3t1hfvUYyjMlyS6XmsXJ0kWMzUwQ5KFUGzwdfYmUNYyW1iQmEsfzK//wfoaTmue9+g3duXqTfXmflyDz9zoCpmRkWV44zTA3N2WM0qhX6gz6VaonNWzd46L5TfPMbX+XBBx7glZde4vS9DzMofP2lF3D09L2UW9N88GNz3Fq7QSfp89ILLzFZX6S1eISwrDi/cgIjJUoFoEFZhTASa3ooMoyg2MAIrAlc7Z27hjtotAJhLcoIjBFY62GK5pewBs9qNNb9Hjtirxje5c9kLcZol5abZQyGCUlumZk9xuz8SQxVstxD29htzjJNnqeEgYclIc37CMo4BUxC4StwiJUmx/Wi2+CZIm3bhb+eO3eOycnWHZ/0QT3084wdjDbGI8sxcLLJwxvQwxZKI59FV8Pdbltz5xp+5/U8Asy01m7Daw+876x1uRgjyX6WZwf2SYc2+9bYg/vp+DUOMRcP+VQ6ZWNhSnEHYHfYNsAUisBWq4XWmm63w3AY3xYocvh39npdrl69SpalHD16lJMnT3L06FGq1ep7JKX/ZLLw/zLHDzt+B5LfDrgeft6dz7+THX2YlSqlolSpocIy9YZlu9vmsY8/xbA/AE+SCcOrF97m2uoqdrhHZDXNiQahVSS9GOMJwlChbUpUKiGkQkrF4tIS+90O1lqCMKRUMI+lcgq9PMvcWjraD7k35rADXFigW1sPmgA6Nzz04INMzyyws93jq1/9BrdubJMkmpdffp2NjQ3+wT/4+xw9NodSPhZcuLHnvHeNNqA05UoZIxK6ccJ2mlL2A1StzNT0DEJ5BMV1m1un8qvW6/S6PYb9PkHg1IFBgVMcvo4Pn/ufF5BVpTm5sCw8+QCD9hp733+B+UxRyp0UPRUuSLmvh/jlKuFkC/a6dLbWCNMB3q1dklNbzD3xONLz6AeSIZbQSEJTGExpgbQSZQxCGlIvx0jp2KNSIn2foBlQataZBGyeYawk0Rmm10NKnJ8vkjzw8IVPCacg9XCNSwV40qmF3XIr8YXEy0CTYcW7P+fRcCCqcfwoKcl0Th7npGkyVoeO/lQqFZTnO6tJKXFK0zsaJ8Y1SK3nkeduzY08Z0nkoZBKkVmLkaOmGWAt2mhWV2+xcuIE0nNWk8YYJNIpGoqQyTzXheJROs/UQrV+eJ9weG6P96HFH09KoiBASBikCSJznwMIh0nIwp5DOLKW0YYsTTl+YoWV5eNI4Vit71eVdbfx/swCDx2QBeoTEywcP44MA5KkYJZZyDKD9Uap844t5zrNsUvZ0tp1nJkdmzuP0WVtCuNx92FK6ZHpHIRFeh5ZkmCM5ku/9gWOH13it37rt1jf7VIKI8qlMv1+v7hRa4RnSI2mVq+zdfMSqlxBSsVMo+YYVdaQ5ylSSfq9nFq9zn63jbWC0PfZ39nFUx7DQcLk7Cx4HqUwYm9nl9DziMpVtNHMHDmC1ZrOzi51P6A/GNDrdfHrdfB85qemESYnHiTUS5NIzydDUKnU8D2fdneXW+vrnJifch3ULIcsRSmL9QKsChmxPUdFBxgKR213LKOO6wgE42DhlAIi3yMKajRrVU6vLPOpj3+M3b02Fy5f4o233uLa9Rtcu3aD9n7HTWhAKVhcmOVDTz7BFz/3eaYnJ+h02qQmx48igjDASA+lBMdPnOBXazVK5RLf+ua3GQyGaDMKC/rLXby73e6YmRqVy1RrFQaDAV4QusXCGDY2Nmi1WqRxzMz0LIPBgLW1Dfe8NCNNMjyVsbmxjh9W6Az6LC2fQEmf3Z1d0mxAgKVeKWF1grYGneeOeWmMKyIR3Frb4HvPPsPnP9vEqgYUxvztzj5Tk7Ps7naoVuoYzycXGVHoEwQheZoivUObjVEHavy5un9LUdDkodh8SEJVRnmBAyf0AJ3GWE+i8xwvDPGkRAnHDtEF29wURQEox7grFmilVNHtFEweXWF5q83i9Eusrq+TGk1mc3JrnOxM+fSHAzzf+U6OvNccnV7dxpZQ0kf5CiOMk+ljUDjrASFBKMvRxTlCX9JqTbCz00bnhrX1TQaJkyB4YYDRxqUf+x5CWR647xwTE2X227sEgWJyZo7pqSnW1zcpVaap1Gp0+z1nQO/5LC0sorOUrc1NlO+/ezL9YrxrjAFVnITPrUgGiaHf3cWzOZ70SNOEvb0O1UaDRq3GMEmxWc7Vi5epTU9w7333cunSJQZxzPPPfo+z99/LyftPs71+i+uXLhKFIY888SSUQ67euMns0hJLy0d5+c3XSfeHBNan2mzR7fTo7mxTn6gj0Chfg6fIhgmB9EnyjGZtApELBjanVKvT3t6l3+tRn5rCCkm51XJynQxKwqPRrCN8ifKdR6odbQKKrquyjDeM5uejNn3f4/0wWIU8kMBnCqQFv2D55sqdY2EtEusM4z3B7uY2//x/+P/xT//5/8hwv0cicnQ5oNKcRZJg+gNEapg/e4J77znFt//oadrDAf/P/+q/4f/4f/8vuHC9zcmHn0JsrPH5L36J9a0NAhMw2EspB5KpyRluvvR9Tt3/CN986WVqjQmUmWZmIqDTfZGZqQVO3/swweQRnu912FpeYK9aR6G5RMJ+0mcj1vzp09/m7/yN38ATDsxyxa3CWoPASZ9G92mJJtCGoayQqcRtqlKLjXfZskPaUY3p6VkIJPfGQ67+8R/wP/2z/x6RK4IgRPcMQuaUMPgiII98PFEBL0HoDInHwGzixRrigO19w7d6u/ydX/07TM7MUt2JefvlF6mfOUYrPEZ9bgGDxghnKfR+ugQ/ieCCE37pMQi/fvkappMSD3MyJL7yWD66hAwVqcoJlEXKEEEVD4kRGqHchr41OQ/kfPyzv04S93j15ecQNkfvbPEn3/w6rXqdkg9TExX2q1Wq1RobNztMTs/xzpVVzj36S2zsdjn5wC+zvbfH1OQcFy5cptUq4fmabjdlMKhy9uyT3ONN4UdVjEwdoGOMuxakKCzBDQiDVdodpfVdaXCIIQ8U65QT5QtbBDcVIUAWCrBAjJuzxmaMwrqEcOfPYp2djtEIUYROWXfv3e8lqKDF8vEzlCszpJkL58h01wVL5qZQ7kRF4m+AMRKlLNjcWQuJg6awsyZwdc1wGKN1zsh3MUmGPPTQgzSbzXd9zmK0Yfs5HyMw1RgHYN/ONHWqM1kohlwYiXZzRMnCYsyRXG5vlL77NQ6/VpIk6CINXQhBkmWFA68bQluE6z2OGU8jKb/loAN5JzPVeW4CuM37KIjqMHg6AnNHsv4oisYgaK/nQoWTJC7qTVUEncqxBcIowGtiYoITJ06wsrLCxMTEmOU7PoZD4N/dbA9+3sbdFAg/6Hy8Fyh/N7n/wWMFAoVAEng+rXKJdpYxjBNyTxBONjDGUC2FfHxhjiDyMWnOcDDAaEvuCxrlxlhWrAIXQDP+HIHQD8jihEZ1EiqG1PeLDAkd+cifAADeg0lEQVRLnKaQJOgsGwP/QrgqV+CY01mWked5YathyHPD1NQU1WqFRqPJL3/6g/z+H/4hm2u7lMslfu1Ln+PosQWkcsu4kkFx7tz1aY1gmMYEuaTuedRLE3SU5q2bl1DbgmNTS3gT0BAl9nY7DLOE6bkWqe9Rr09SkZosG7K10Ub5koWFBae+uwNEfX9z97bWyvt4/E/miIygElRI6g2Of/pTvLM9oP/2NUIhofDOFBgXzDjYJ/I0taUpVDtid22LfJhSUinrZovWo08wM32Mnh8xVB5tYoQFJRQeHqHwEQIyHIgrcYrJVILxHMknEAqFQnkeQehjqiXMyM7OGDzrGut9YRDS4mFRRoMyKCFQmYeUviMwGUEuizXzENgpxAHJw1q395YIrLb0s4zM5Cjfp1StUiuXCXwf3/fJtcZYyYj7lucOc0vz2IGORrtMCumTKw9PKgf2Cu0ASEnBSnXNtTxPCxarh5UWH8HK8WMEBSM0Gw5dg8PzEQU2oK1x/qbWusBYwJMKbawDeIWzDhyt6VI4f1WtDbKwErAY/MBdqT1ryRHj8Flrc2xuCrcYt7fIsoxq4PHhJx6l5Hk43OzHy055X882xrrFwEIpLHH0+DK1RpMkzR2AOuoqZ+PouvGNL8sykiTBt944gGYkybgbRXl0woQxY7+aURJ4kiTs7u5y9txp/tP/zT/id/71l3n5ldcIg5CRHF7rHE8I4tzJqI2UyMEAz/dJZ2fGJ16bHKnchNpv72GsJQxC0uEABeRZShz3KJUr+EHg/FADH08I+v0eYRQy6PUI/MB5rQYBpcgFVPnWIrRhv9dhamqSeOCShYMCFPZFiPQ8SpOT9HLD/iCm6kuUyZE2Q5kyKhSus+tJl3A5AkuF86QVdgR8uQtJMPr79k7f4a+lAOlLpmdaTM08yqOPPUSSZvR6ffr9Ad1OF2Ms1Uad6alJmvU60lr6vS67OztsbW2PPVWtMeTWsdKmp2b4/Oe+SK3a4Ktf/Sp77Q55/m7Jx7/vIYRgZ2eHVqs19nEKgoDd3V0EEAQ+5VIJcF2OWq0GwOzsrCsSC8Zdu93GaIMP+L7H+to6vW6PUhAhlaE3GI6PTRVdKBmEJEnigPosw1eKb3/7O5w9e5bTp+5HCOu6TtIfcSqI4z5KgbEeQeAWNmklXsHQHA05BspH86Do0MDY00cIgVTuerEYBv0+xhji4RClPAcYFiFkolh59aHEzQPm4egmWoCsAnw/4szZc5w6fYbL166RWY0fhSihyNMcYTN830cbTRiGY6nWyOR6JONzQIkhTVKQgjzLyLPcSQfShFq9QrVW5TOf/jilUNHr9clz7Vi2ScrG5jad7oC1tXW2trYIojKTkxMsHz/CyvIxup02YegTBQFKSvrdPnEcU61WqFYrvPnmm8RxTKXiLCD2223KUcn5yf5ivK8himvEFqGCEsGg33cJpAqs1W4uG02lXKW93+Z7zzzLL33yk5x74D7efuUN5hdm2dnfY+n4MR559DG+/Hu/x1/7G7/B8fNnqExNUDMef/h7X+bJz3yCz37hC7z08stML87z4GOPEtXKvPbqq1hyzj90LzevXeVs4zzGs3ga8jjhjRdf5pGnPoTKDcMsQeegrMQKS6fXpbPfQVbKeJUyyvPQ1qKtplKtOv9cKZzMhNs3BE4Icoht+HOyt/pRQbO7brgKtpEpGjUjcJ4C/BLSxd0I4bre77zxDv+H/+v/mZs3bjHMhyQqZ3phlqV7TnG9myCGGYuLs3zyr32J0vIK2g/xF8/zx//sv6d9/Rbf+7dfZeLkCoMw4uqbq+j+s5x+/Dy/98df4Yuf+xK//OineOZr36Fz8QofffA+pkK43ImR0RG6W1VumsscPXGMZilgq1Hh2sQs+8Kg5YBJYSmVPdTA48n7H+P73/o2n/nUx5ibKBcMqwCsLBpgzk4FKwtgS5AVDELfCpABVeUT7A/x2oZ0UqHTmIbns4DHpdVtKrUa5VaTdL/HIOnTak5QrU9xZWufybMPI+Nr3F+9ytFyyiCG514ZcPzIce49aUm0T5cVOuvPsLNzleWjJznjzfD6y9d469lv8ff/9t9EoYmygKH//mX+7zX+qoAHd8vyCuZ8yoWLL4DokyQpUikmpydYXFpCycABi7gEWEQOOG9kYT3AQwjP/RxNEPo8/MhHkeT099v02inf/sbX2d7eZnp6msnJCb7wxUfZ7lxmc0/z2BMfBuVz/EwZawXS0ygZcua+pwj8Kp7ykb4AqUFY53uqjFunLIXv6QFz2YHwdlznClShWBoxcQ8oua4GEGNPZ+u2HO5YpCmCHzRW2ILViiM3aKeAcRs1g7UOpNMmZbfXI0t9mpOnmZ1bJtM4P35hyHMLRjDoD/A8r9iEHdR/I4uOEQAmbls8D+r+4XCIVM42Kctyzp8/T71ev+3zvZtE/Wd5/CDp/WikaYrnObXT4ecIoQqyyoFE0vlRSkQRQqqEu8+Nvl+8wFiKP/o3I3DxEBM1z3PnTT+yhhCFyrB4jhl9D7eJvauXZqHssliMNmMZtTUWcSh34zCgKqXLj6hWqwDERShsHMeFvZQ7Dmch5c7P7u4u6+vrBEHA+fPnOXHiBHNzc4Rh+J5z6udhfv15x2Gw/jDbV4hRt+Zg3I2JOvr6MLN4RN7A3l4zlCsVOl2nAt3b2+PYsWPsttuorJjXJsD3fMqtSYaDofN+tAdAUxzHhGFIHMdjO4qJiUkXPhOEeJ5HHA9RyhE5KOZ/bNweDkSxVt7JspVjiXSWGRcy5HsEYcTpUydoNP8mly9dYfn4Mc6dO4MdheIIe6h+dP6siRHYPCALLH2Vo+yQi2tb7Cb7NPyE/XYH9DzdqE41qFCLIrRVVKRH2vTJkKjcp97w6fW3uX79OvPz81Sr1fEe7E5m8HvP68NduZ/eoYWlUqpQK9fZ6fR48Auf5237FdYvXqKFJcBipUAbQSk3ZJ0BfeUxPT9LfaLF5toa7dVV6umAwU5OdXmV5rlzRHNzdCXEuVNq5NI5kES5JLIG4wlSz5AL48z3jMDmGZlUgEDaPp6WSCORKLTGEeKwLiAJi/QgEQZhDX4uySRoqQmVIhTunh0gQEoyfXANHXzGbt1TSjEcDPGVT7PZoFQpuT2NlGO7iuFw6PbhBrQ2KOVjrPMFDsKgKCdcQDoWpwAVhzyqCwW6KTytpRC89cabzMzMsrCwCMYWe2tnDWq1ZWtri2PHl8dNNpeDIsZKdls0LKSUeMo7RMRShd0oCGEY2W9Ya6FYfzwZYBTkqXaWH9Zl0EjhZP1xnGCx+J6iUa3yyY8+xfzcLOjc1T0/5tR/fzJ/Y9AIpFJMz83Tmpwhz92ipZRCeQUtGIcsG+tMaLXWCBx1d7Q4jSQeI8nunV0TW1CDxzf+Qz93CY0Bg8GAer3O3/+P/x7PP/8Cv/d7X8bojDTP8ZTCGo2nFHmWjVOafSGoVyts7mwXfkwOeI2HfQcE5ZosjV1uqRAo5eEJGPS6BGFIpVohzzVJmiF9jyxzN+okTmjU6xitiUolwNKs1TDWMDE1gR+FVJQi8DwUgnKpgrE5lVqV/jCm1+uzY/fRvqVc8dFZCW0smREkgaVkfFQho3aMBeEKIRwzTFm3SEuKgoeDQujw3BBSIhglpLvC3JMSLwqpRCFiasJ9PsiCjWXQaUKv36e95wqRkUdmmqQIKTDakMQxSjqT4aeeeoogCPnKV/+ItbX1Mcj+lzVmZ2bo9nr0+30GAwqzY0mjVnM3Vj+k2WwSx85PaX9/n8nJSSqVCp1Oh8FgQLVSdZ+tzSmXS8hMYa0mS1JKUYS2KcITxHGMpxRSuIS6NE/Hhs1SKdIsR2vDnzz9NHOzR2g0WlgjCDznrVouh3S6bfY7bRYWFlxaY27wPLBGOYZAsfi5i+Z276vDBa7n+WS5dr6ifkCe5/T7AyzOXyQIw8IGYMROOdDGHVx7tviv6HKNNm/F/+v1BufvuY+nv/51JIa0KFRHi+moOHeeJtpdW3FOEARjNoEUAqxxJtPGoopmRhq763E4GDA900LrlDSx6HxIFEZ4nkeUSuZmTlGqVLlx4yZJkiKUR6vZpF6rkiQxSV9g8xy/VGYwGCIQHD1yjNXVVYQQTExMFP4sgiSJybIMr1obM7J/Me4+3lV8FWu4FBarM5JBD1U0dBACIyyNiQa6M6QxPcmRo0vsb23RGfZZXj5GdaLBrdVbvPjs93j4icf57Be+gPA8ersdOp0ejXvPs3L+DOuXrtKamODMuTNo4JXvP0/r4x/l1NmT3Lh8mWNnTzAbLWKVBSWwueba629z5uQpMq3x/YDUaHIL5cLjtVqrEZRK+KWIqF5zdYK1KOURRsXGSrpu5+ia+HkfdwNTf5SNpoVx0IgcBYoI4SwaLAhPgTZsb+3wxuWL/NZv/zaXLlzixvoqejBE5ZrQl/R2drn12kU6JIihpRznrK1tsnzuXkSpxuNf+hLHZyb58j/+b9m8doXPPnofr12/gPWHXL5ylbevX+GxD9zPjesXeeLBx/ndN17GK/k8+9xzfPT0/bz4jW/Rmwp4+vvXaUeL3Fxb4/TWNtNHTuHZlJb1yYQkiod0bq2SJgN2L99gY6fHc29f4jNPPoBnRcHeLhiGYwGRLRiqkkwFLsEdjVaSuZkGMyJjd22NxmSdU4N9Tu8OiFNYufdBPjU9wWDQYbJVZXtnm72dLm9t7lH90AeZO3aMpa2bfGoWpqVCRRWq3pA//NoFjh9/iPNHDKdPCyYmXmbHXmZ9/RkWgw9zbu4ME80WU9NzCHwSv4+LG/rRx18lCDHaDloDvf098ryNUilSQlQKWDy2QLVWxxkjCaRwjE+n8Sg2DkUd5Ia7DwrhIT3AuBDRpz76cdJM8/Wvf40Yzb2PPcrcseNUJ2cplaq0WtN4QRkhA5A+RsaO32S9Agg1jp0iJKYIGLWAtPp2likUgOqhjbgt3hMH9+jR4w7eslNpOWzq0CbZOhbWCDBFO5aENUUSd/G+UmOIY8MgThnEKZXWMeYn5pGiQmYU2o78OQ15pkljB1AoNQp4eLd3ojukA+/K0UZfSkGn0x03HQDm5uaYnp7+BaD1A8bonB6W9SdJcghMLT6DcaO8qCeVRHoOvJJC4IlD+yx54Bwni8YpHJ6SwpEGOGjkWyFckJk4eF/WWmyWgRxxnW0h3Sx+PgJWjR3PI1uAp6KgqupDQOqo7q3VapRKpTH4Pigsm0bnwklBDwgG/X6fGzdukCQJS0tLnD59miNHjlCpVP4cjL1fDDgMJorb/j26jo29fU9xN2uAu7FUx79nFKAnDqwbwjBEa8NgMCDPcwLfx+Q5wvdBF69tLGEYMRwOMcaBqA5wD1HKc0HLgFJeoVx0exffjyiVIsIocEG6o6Rwa0mFY/dZW7TQx+/7YP0ahfVIpfA8iVJQrZaoVI9x6sSK85uUAMVc58D7WOscbazLKREG8Ejw2NxaZWN3G+21mVwwLNbnMdkMeZ7RWX2TWr2EjI7j2YBut4MNPaRShFWFsWWMseNzFUUlSqXoXZ/hDx4//QVvL7CUcsWpE+dYv7VJUq5z8nOf5oWnv8r6hUvMDFOiIuRbY8nRpHt7aJMzOTfP7NwcQXuP9m4b2b4CvTb9zas0T52kunCMWqVG7isSzyf1I1Kh0FJgrCDTrhEqjERZ19BUtvCuzvyDerfY1mON8z2naLdbixUSI3w8oRyG4rm7ciwN2lcEysMzAlXUlKP7+4Ea1IUSCgGTM5Pj0KU0TQmKYGZrLUV4igt+932H5xUKFVF4+I+auaLQrGANSkK7vcfG5gYnT50qSF0ObD1z+iyBHyAtbO3s0Go5XMlai1SKpSNHHAhb1Fdj/OAQWjW2p5FO7n+4ZjBG39aUM9ogDHhIUAqZS7qdDoNhRimqUKmUyWxOlsZ0Oh2M1pw9dZIPfeAJzp44jjDFubcwygH5Ucf7AlO11hgraLXqHDtxChWUSIuOjJQQFsnJSUFMFUBa0OFHxVKSuEVHKjm+CTsp8YHx+YE83SXPj090cTIdtTcnikKSOMbzNU994HFmp6f4l7/9O2xt7ZBp45JLtUP3hQGrc4Z5xv7uNjbPybMU5QdEYUCWpCjPZzB0XV0lPYIwIEsTfD9A25Q8TchSNwmH8RClPcIoAgRB4DMKzRnJSUZfG+0W3HJYQlhBoHx3LoWm3++RJTmBF7A/GLLTXqNZ9qjU65RbCaVaQq1aIQ8CAj/AD4ICVJUIU/haSIkRsgCs795VHdP8bTFhjTgowgu/QxcIJMa+VcbmZGlCEse0d/e4ceMGIFhYWEIXnTkpFMp3XqFJkpDkGQjJfffdj1IeX/nqV7l+48aPNTn/vOOtN9+kVC4zOTnJYNCjVCoxWQBoaZKS5zm1Wp323h6mYD5rrdnf33fhSZ6HX3S842GP9fV1EIJSJaRcKmG0odPpsHD0KMY6cD/PXdPA9308zyNNC1DVUxhjeevC27zx+ss8+ugTKBUR+JJ4GDM9M8XFK5doNScKX1CNVApr1cHiiGLkTWmKxHSKhc6tRe66iEolVJYTpwlxEmPyjMEwplQuUW80iKII5anDQqzi/3cDyQ5+7uR+YK0AoThz5ixTk9Osba2jhCLLU5QvSfMchSLLczePtBmzB0bpmOCOoRIF5EMH/AZhCSUUJjcYk5FmCWmW0u108aRhZmYGTykG/QESwX57DyUlpcAjCgrPoSxmd7tPs9FgemqCfn9Au91Ga8PU5BS7u7tUahV2dnYIw5C93TbHjx9nb2ePQW9IOUocm/YX4weOOzceQkmwmiwZksVDIq9gKnhubYnKwf+fvT8PsizL7/uwzznnbm9/L/eszNqX7qpeZ3q6p3umZ8WQMwApgYRIkDYpiEELsh0OSVbYDlly0A7/aUc4FGIEGQrLtoKLIAq0RYEgKKyzAph9enq6e3qvvTKrcn37u8tZ/Me592VWdQNoEDPDmWGfiOzMznqZ+d59557zO9/fd+HmrdsQK3qtJs1ui8VT63zj81/h4hOP8MiTT7Bz/RbTyYQ7W1tc6Vzm4uWH+J3f+J9IVMBDjz/K7Zs3EU7y6osv89wnnufpJ5/k+ltvsnJilfMXzyKbMUHLN8KUlcwGI25fvcaFR6/gpKKwFkSAkv5elcahohChJLlzNMIQhETrjHqt4eU/Ar9HOTtnSrybNO3fpPFusrz3cj3e9TDlCahluKJvBGrh55MtCn79n/2P/Hf/7a9w9fYNdvp98lwjRYASjqjdJOq2+fgv/s958pmP8703v8NX/ut/THpvm8//w3/E//GxJ4lOdzmoNzn91HP8P/6fH+LWzdf42tde4OYbN4ljb8+yqBxxM+Htfs4vPHQB+52vIc9s8j/+5u/wMyc/wNMLm/zTV7/MpStPsLM94jDo89UXXuTPP/wY9aUlpi6kk2V03rjK5//RP+DV/S3iKORUs83NN95CP/GkPwTaDCErkE4BpgTuJCAxQgGayEDhDGMx5nN/7kO8/KVv8vSFU1zJxqhbd7k1mvD2zj2m0ym7e9uoy5c4e+UxfvOf/ToXPvFpwvUTDK5+j0c6dYqVx9lz92gx5kNPrfP69TG/841tbPsy+tZVTvVHXDi9RFOFpNsvsjvuMwsmRLUQa2fkShIc6y39q8z7f/33ikGIgLt3t8imE+pJTC0JCQJFq1lHqRAlZbk3uTkI6Y/NFZgqcOQgDM4VSGlwLgdyjEtREXQWW8T1mN29A777wvfY3DjJ+toJAiXJ86lPpFYFUgQIYqQIcKiyCQqUElKJ8tinqwDS+UZcNsiZW2OIssnpbAVqPACozodAuGN2AeVe7vuZwtv6OoG0ogRTwQrvmDvJCw4HMxA12p0zLG8uYmniyDFCY2yOMQ5nJU772j6OQ8CzFY+e5/2ASTWOAzBCCMbjsWfAGo0zjo2NDTY21t8ByrwPfN0/qutSeaY2Gg3yPC9BIIcQqnrgfP8Wyr9nUvuDo5IKLbyUsqTxzJHT6qQghPcHB//eFVp7D37piSf+WCHuO+QKBNpaz8bnATaiPWI0UgGr9miftdV+e2zuNJtNms0mxhgmk4m3GjCG0WjkfS/Ls6L/Hf7AvbW1xe7uLt1ulytXrnDx4kUWFxffD5f6AYzq2s09eKt1yTmM0e/6+OMsuuOs1vlcKB8XFDlKKoIgol6vk+cFRV6wX6oAJpOJl+KHIaIEnKIoZjqdkqYpb731Fo899hhaa8bjMaPRmDT11g97e3s4Z3niiSdIksT7jAaSOI4I45gkSZjWaqTTKePx2AO05v4AuCMA2O8fSnm1qwq8XZmsQp2FLZWcogS3xDwfJlCKQEkMGsixRYP9wSE37t1FBRPOrqWsNtcwtoMSAcvju7id73Jwc8hs/UOIS88ho6Zf1y0o7a1VhBDcunWLwWDAysoKKyvLLCws/BjsyT/a4Rysra2zsr7J9o0bNFYXuPLX/zJbv/V5Jr//bcQsJ4umqCRAOkNLhOjBmLvja8QLCyyf2GDl5Bnu3LvD7r1buN2bZHfv0Vu+Tq27SBpJiqU2bmOV1uIyMtUEtQZaxXMff+ZneesrPh1jFOjQYqTGoZFWl3WHpBDgLZaqZm5AZAWRtbhQYJCownumEpTzzjHHyI6rwYUQLCwukmbZ3MKkVqsRhYEPZQu9f3pudFk++HwSf+1KBQA+SHzrzjbra+tI6XOOEBBHEbU48SFUUkL5+FqclDiYYWtrmziuUa/X5wQ+qbyixlmHMxanjmqBivFaKdIp1QmF1t52raopyudYYVhKeCKGcjCbztjb3eVgr0+z2fQWLqEPujZ5xpWHH+Ln/62/wGKnhRIOWUZwW+EJDn+W8Z7AVG0sUsLGqTPUmm3ywnipudDgLPkMcl2U6Y7eyyDP8/nFiSIvo02SmKIoSNO0TICswi0cojJJLzflUAUY7H2FlFJ+0ZPS8zwKrSmylAsXzvK/+0/+Y77+zW/x8svf5+2r19DGvwk4fECJtWzfuknSbGGMQWc50hiwjvEsRRvvlWWFRSovmc7yFGs9IDrb3yVJagRBQBh6eYGnEh8lkVrncIWnkAvhywinHZGKSXVOmme0mnVMkdLpdkmtZjQdMcNw89oNotmQpU6PpbUTdBZ79HotavUGSb1BUquRJDXiJJkzAJRSOOmTtU1Z1FSl9bzEdnjfLQcIz7w4CkQtJTYlc9AaL2/I0hHD4ZCD/X0ODg44ODhgc/MUo5G3AbAIgjAgCDzAXK/Xym5IgTKGK48+QrPd4n/6zd/8M03OP+2o12qeWZpnNOp14jhmd2eHVqtFq9lkf3+f0WDIcDCi1+siA0Ucx/T7fVqtFnmec3d7G6UUOs9I6k3COGYwPKTdbDObTOl2OgwHXipsHdRrddLMy4uyLCMMQ+IowTjt/eeE4Itf+jznzp5naenE3GA9S1OajTrT6YR2K5tLr3xn/ei2rDbfip0iqhJX+IYBwv++OFE4AVmWkU6nCClot9vUGw2iOKaKsJp3YSiZB39C9xjhZbfWWhaWl3nqmWf4jd/4F4AjiWveViP0MoYgCOZsGleCzd5+QILwxdMkzZAqIJIehFcyIAhDjHEgHWfOnKZWbzHsH3Lnzj3ikklaSxI6nQW2tu6xurri55ryB7HKv1OFIY1mi62tLYIg4NqN6/QWekwmE9bW1nDWMR3PuHPrNnmes7a2xmyWMhyOfoiz8id7HD/QHh/WWQIgT1MEpdWL8psi0uKQnLlwFgsU4wm/9bu/yef+4l/kqWc+xDe++20ee/Qxrl+/xnNnNtnc3OAPvvRlnvv0x/nIpz/O3r0dXvnuC9zbvsunP/fnuX7jKm+/+QbdlUU6wQKNZg1RD9DSIAvLYP+A5UaHr33xD9g8cxoXRRwe9AlrdaJIQtkMlFaQaU3QbFBvNH0KujGEUUgUhwjFXFkhEGUD48cBHPrxGu/1erwbACJLy2/NETteOoEwsL+3z/e+/xoffObDbG/vEAcZMhAs1Xs8fOUKvfMneeizn2J25gxbrkbz1AV+prHCd/6bv0f/9h1uvfASjf0Rd/b3yPSML1x7neztq+TTCUsNTU0FbFw4TRRGfOetXS5/7i9xr7PMytImcdxmd2OB/89v/zp/7cOf4W+KZ/nHX/xNRLvLWqvJcGfAyu6MfM0ilGV9YIiWezzxuQ8z+vK3WHvmeWrjPsrucLh3jfr6RYQSOAx+N1bl15VvoEVZ75WlrEBpiwkjnvno0zz8O9+k/rXvMju3xotvvM7buzusNBs8c+4ck5MLFK02L3/v+/yNv/lLyGaNcDKl9aFn6dYDBvmAdjIkzid06lP+g19+hn/5+1ex8Vn68QcIijeY3bjG2moAi3vs74x4/INPoQIF+YRAx97/0D3AaHyX8eMIRvidQFPkBXs724RKosKYXrdJP52Szwqymca2NM6WzUonwCrP1EDgQW+HcAXOaV9w2wxcgS5SrNU4W1AUGbrQmMzx6kuvEjjJxz72PKfPnCIOwOUGF4Y4K0HVQIYIGeOc9FZJ1S3gvCeph69kRZPwTcyyRjtilnqQSVTFnTu6H+d7/PzfHIJKpmqPGCbW4WyAcxJHjhWOcZpzOBjjREQYN1ndOItUNYxTaKd8sq4TWGsodEGWarCSKIyoN2po7Wvld/PJfzdQtJKgplmKNjnWWbI84/z585w4sX4f2Fr97I/jfPvXNebEk5KMMh6PaTabtNtt+v2+l3By5C0qS4IHxssZZQmCVUGix5mp94Uw2UqKybxwnJ8XrEVYX+Pqqmku7t8fhDuye7DGzgHSClCtwFTgPim/EIJ6vU69XidJEvI8ZzAYMJ1O50SIqt6uEqu9RZ5lMBiwtbWFlJJHHnmEixcvsra2Vp45359Df5bxbtdPlYzOeXPHHYHn7+Xnj4OqnvGpkVKhpAdlrbUkccxwMGR5aZlABZjCn92NMZ7EkmUYbYmjGlcuP4pSAfsHO2itabZarCyvopTi1MnTvPjiCwwGA3q9LuAIg5AoCgnjmEajwXg8ZhrHRFHEYDBgUlqmldzB+563B7S87NhZg1CqtA0orTSkT0I/Pq+PFLcWicO6GpPpLjfvXCeXms3WkBP1HnawjotaNA6vEr3xhzTzmwStLrN6wt7OLpkMWV1aJhCSQijAB+qeO3eO4XBImqbs7x8A/izo/zYcb779NDaq6tqDY6FSfOgDH+RfbG8xOBiyvtjhkZ/7OW4urXLj975IY3BIK9WEUjBTBVkQEuOQ+wMOBzMWT6xz8tw5Vk6fZr8/4Pqt22ztvc5qu8f6whLR9gGjl68yTSK6vR719ROkzQZhbxFdr5MrhU0CslCghSUUXkKP8cosoTXK2XkTSwjh6w5rMSiKKEI5SeAEVnr2f2Q84UNLgyubYNX9VzUyrbW0223PRC3Z3d5GwxJIhS00URiR6tyHRdkq98UShTEWMVccOOvYubvDibV1PBiKJwjW65w6fRrnHLdu30IKyYkTGyVe4cq199F5w81ojStJEwIQVhDHkQ+iL++NCkebNyvwzT4bHNUM1ZnBq121D9iUDu2gyDXGaB5/7BE+8bFPEkqFtoVvcghBEIQs9rrU4hhpNUpUzvHi2Me/+nhvzFQHJ05s0FlYYDiZYozDluawOMtkNODu3XssLi6SJDWEqKQ+jiSOGfUHJLWEZq3OneFdDg4OSvq+9zRQQeAXIllt8Eddn+PdLF9gSeIoQuNl+lIA1tBo1PjMz3yaj37kI9y8dYsXv/sSr772GkYbDvYPSfOc4eGh/x0ILwsYDjBae0PhOCJLPegKbYIwAuElLYEKyGZTlACr4xL59wybWZZRqKDs/nqJvKcoQxTFOAfj2QRXpo2ZImOt02al2SBTjl0J2zt32e33Obj2NjUhSOKE1mKHlfVFut0F2u0u7U6PdrtDq92m3WrRaDao12tEUUQQKIQKPSuwetPK+ts5hzBV80EdsRSofLIMhdZkacpgMKR/2GdwsMtwOOTw8JDXXnuNNMvY3LjFyZOnSGp1Wq0O3V6XKDEoFVIURybvzjmElJw+dZpf+Pm/9GeanH/a0W23qTfqFEXB/v4+6+vrJUCsGeZD1tbWEEKwsrLipRCzFAEsLS0xmUyoxQn9mZd/x2HgUyVrNbJ0wnQ8Ic8yGq06/fGU0WiM0YZJPiapJV6CEpXm4lJQjxpkmfdx2t7e4qtf+0M++7l/iyiqez+T2YzF3iJ37tyZ+5oarbBhUHbWjweJCY7vdVVdWx1IqkImDiOKPCfLMlqtFvVGo+zywBxpB4SrCp9qMTnyPzreYRYV29mWlgMi4BMf/yRf/vKXmYxH2LJZYowlimIwem7tYa0vcrTxIRhCeP9UJQPvDasUFkcShRR5hpAeHI7iiDhJcA627+7Q7XZZXV5m6+42vW4XKSU7u3uc2NjkrbfeIgxDkiShP5hQq9cYDvp0ev5xSMHe3h4Xzp8jkIrDfp+11VVGoxF5ntPv++7VdPKj9fb9SRsPFlp+aZFYk5OnMzDWS1OcDy7BWITychWEoHdimWc/8hyz8Zg7d+9y8tRJFrsdFhd7bN2+zUKvx0OPXSZu1Pjmiy9y+sJZFlYXGd3dIU2nXHz8sgfclWOpvYJFo6VFleLXvZt3ePHF38MYy+bPPE/hIK43wQhMWszndOEEg+mU5cVFgjhB27IZpgJUoMq9oXyNx/adP+l6/LSPPwuY/ODPSgdGgJMemME6rl+/zu/+zu/xyivfJ3eGl196iemgj7J+vYgkvPrdFyhe+Q4vvPI9nvkbf4ONZz8OrsHyU8/w+HqXV37j85hxzo0vfJk3Xv46QzVDLaxiClhq9nAkqPYqM7lC1BNccH1Wp7fYGl3huV/8RW5+9Q8598QjvHZji//iX/wTPvvIB/lrf+7f4aW9t/nqS9/i4mPPMFtqkCvQgWTcbjFshITt5+i9vcVnP3KJxvhlTqQvcv3ar3Py5C9jTBNkCOU8PZKQCwSa2ApyKcmVwBlBN+oyE/Af/62/yX/5X/x9Xn9ZsV/A5ac/SFRMuXLlMidOr5F2OvxS7yTNuEW7F6NkRkaMkwGxtVgrymCvDKMHPP2pQ27cGTLRHQaDq/T3v4gav0YwivjKN7b52V96HmEdgtA/RVcCvmX3zYN+795U+XEb3s8Z+oeHHBzcQypDGEs67SaD2Yxb17c4s75DLVF0ep05cCmc9EBgGaQAUF0CSoUTziGdZ3CY3AczBEohlJd73rjxNgsLHZqNOklUQ4nI79MCpEoRsgDp7WlwpSOzKAmq1R+kOjAc8xJ21RtBtfEfdcrnPzd/owBbAlde4SWExQkPEOfaN7qcCEizjGE+YTAcE6gm7d4ZkloXIZMqsxrhDM7maD0BAmYzL30NZEgQSazTaOP8nl56dMIDgNoxIPT45yzPyPMU8E3gs2fPsLm54c8V3M8Ge3/cP47fh7Vajd3dXUajEa1Wi25ZJ41Gk/kBWx5j/jl5dHiXwoeSypK1yn3vFf4eEAJxzLKrkuhXDxLCB+tZZ73c3/m5al3lpXn0vQqUPc5ONFqjhMRYQxCENJoNnzCt1NzvtJIuG1OGkJaqMlcRQpxlOs3Y2tpiOBxy+vRpLl26xKlTp6jVanPA4f3xwxnH71UhBOSehXkcUD2u8HEPzIEKLKlAeofwIT/l+UUphTGG3d0dlpeXGY8npGlGnPig66A8hwI0Gg2s88o2W847Va5NSkkuXLzI/v4eaZpSr9cAby0QRZEPqA1DanHs/YCdm6sN502qd5Ekz8FWUa13zrMRy9skSRKm0+n9a5pzgGSmC27v3WFcpNR6BUsLbVx+mjhcQ+7dxL35B4TDO8yaXfTmM9jlU7RsQl3D9772+ywsLbN89hxKHPkit1otms0mWZYxGAy4efMmtVqNzc0NarXkvnX6p+2+CKykUCCspdtq8IEnHuOFL/w+050Rs8UGnY99hKfOnuX257/I8K1r6PGUermHZWjvf5tnFDenpKNdVs9c5JEPfYwzz8Ltu7fYfv0Nrt66TTjKaCcJNpLciUA0vo9rJESdDp2VFbpra7DYoejWmYQCZIRyitBKRFYyioXyZ2tVGjc6AYSkSpG6gEwpnJCEGjCGSVyQJYaGEcQ2wAqJFPg6RIg5gTEMQx+2VOaWCAFKCKzxFpjOFL69by2iJAx+/5VXeOKJJ+ePwTmUUnzwA0/inMfl3nz7Lc6ePeeJJsIDvI1mu7ToqPalirXu57oFCqMJVVhRCe7T01RNVWv9+d+aaq/yj5GlB7ujBFGtKRvavqaRwuCsJtc5xhpWV5Y5e3qTUCkfyqgoc36Yk9KAo7qK6in/2Ugz7wlMrTfbnL/00ByxljLw1PU0I82mfgGKQpI4wWgPrIUlg7Oi7NZqMUVecG/7Lndu36bb7RCW1N1qAUOVKXdSYY33jjKlbN6bnSv/JhiDU4pQhfOujxRQFBlxHPLEY4/wxGOPIaWiyDJe+f6rDIYjpmnGzu4eL738MoGEMJSkxvrk+dx6/1GtsdkUYwqECgicT1urBxKbpYjAewMN88zLq1WAEpLpNPMyfCVRQQQanFSEtYRROqNwBSYPSIqIT33mEzx6/izOWUbW8uu//VsMd+6wGIWoPKPIZozTCS9//y6zaU6oIpJajW53gYWFHmurq2xsbrC8vES326bVbBHEHsR2ziKFT0ytFlYpSt+iEqD2ElaHzj1LeDQece/ePe7e22FnZ4e97V0Ggz67u7ukWcb62hq7u7s4B72FRRqNJs4abOnTORgM5gbfUILhWJaWFv5Mk/NPO3SeM3G+K3P+3DmKouDE+jrj0YggjMimM/bu7XjZexgQRRHj0aQMTlDU63U6nY4HWKTAobi7tc1kMmR9ZZ1GUuP6rev0x1OyLCeJ6+TZjKJiYEpJUeQY482djdYIIZGB5Gvf+BqPPv4k585dIgxDJpOMjgqJwpjpZOzlJkqgS1lIJaOSUuLKcIB5KJWbE1TmMhtflPjuqC5y6otdgqoTL44E/QJ8cNADjNTjDFWBAFkyYUVpzGwlzlpOnNjg8cce5w++8mWMNcjAd6/yPEOpwC9WAqQMKHQ+L0icc36BN16So21BkkTEsUKJACkd3e4CK0sLWKtZW19heWWR7a07vPrGKywtLTMYD7HWd7Bu3rrB+fMX2NvbY39/n3q9VlpshEwnU+r1OlJKnnj8cUbDMVb7oqhWqxFHESvLywz6A7TW9N4lLfj98e6jkvA5/KGqKHLvAOgcSohy0wWMQYeCsAw1WdpYRVpodlvcvHadQhgeeepxbr51FaEEb954m3CxyeWHLzIaDJAnmjz+8Q8TRwHICBkotDNIZzHC/73QgkGwuLzEy4M+lx5/FJKYPCtAJahy1hfWEAhBoSTLG5vU6w2Mq16LT3hV7yLxeJ+R6seD68Sf9Jh3e2z170W5hoUWbl67yT/4lX/MH3z3BdIs4+7NW+STWZkKnhMqx9PPfZiVk2f50ue/TBIE6Js3+O7/+x/ymec/wVU9ZRqHFJ0T/PzP/xLF4YjGR57lc3/9M/zz/+FX2Ro5Pvhv/yKjaU5zpgkciL2bXFh+g7WnAn73t79A6+QV7tzV6FBxcWGdVtTEPqz4py+8jLuV8qmPfJC//bH/Ja8qxVu9mDQKmISOwKS0UkcQbvLkB59m56u/yvMP7bEmb3JjPGCabhMHlz1IJ4p5CenDjRxOFj4MEEkhHU4JEhVgUJx/ZJP/9f/23+Pv/1f/DWoyQPZv8dGf+TSPfvpZllcXvH99ERHZCG00BkHdGCQjIKEQAVo5jAhQapkoXub8FQitw8iHGU0+zLUXfoNvvvDbhN1zPPXMJ3BmhkOhkSjhjlhoxxjEPwmHLucgz2EyyVhcXmRv2wcONJsJ7WmbdJJy49o1nEzZ0CdI6m3m+YMViIxXXAk8m4MyhMo5C1bjjKYoMrJ05t9VEYDwqotr12+wtr5Od6FHEEsCvK2IQiFsgJS+9pIyQMpyLlRgaDVEFVo215JQAa3OGbA+Cbiq8SpgVYijg4pUEq/g87WfxctvtXNMJlPG4xRQuFqP5ZNnUCLGGYk20kv/ncXoAut8loB2GiUihLO+zhOWQnuv86LwdirWHdUM85fyLnPGA30j8iJDKYm1muXlRdbX1+YHsPfHex+1Wg2EYDQakaYpnU6HbrdLo9Gi3+8zmU7nMuOqHrTlfFMlM7VqJLoKDKIyA3Pzc2a1X0LlTVqGgxxTFZpjao65RVv5M6JsTsyZqiUTNQgC6vU6zWaTKIox1jAtZdZp6sH26rlX4I/Weg5gGaM5PDxka2uLdrvNs88+y4ULF+ag8vvjhzuq9+W+kLNjYF31PleP88fQBxr0x4DVufSfI+/bMAgIAsVoNKLT6RBFIdPplFwXBCqi2Yw8uI4jyzKcs4hAzi3svHe5J6V0ux3SdMb+/j612ok54F+twXEUEZTzJs/zORt6Dt67I0/9LMsIgpAkcfPmmyvDeJy1uJJhr5QiSZK5BYdzDikUmoLbO29xbzAhDxwtEYM+xVS3iIsh4u0/pD68zuF+zi5nqLcegaBNU8Erf/ANbn//+3zXTjj9yCN89NmPAaIE0MqcCulBtmazyWg04pVXXmFzc3OeH/ETsKX/qYcRDmXBSoHGcunSJabDEX/4zW/QHMKiDqgtnuDkX/srXH/rNQ6+9i3yF6+xbgJmkeGenZHEimZmCA4z9jLH7HBKe3Od85tLnH/+w/T3+hweHLC7d4+DnbvQn9IYB9TLBtU4CMkbTcJel3h5iThOELUaIgwprCbNve1gqAKE8RJ5rCMMElzYJFxaoLuxTrrUZRZFkCvCQJDpHGPB2hDjKAl8yjO5VUi91kCbYh5KaMr7pygsMoowWvPd73yHxx97lKReQ1qLFxVKNk5soKQkyzNG/RFLi0tobRB42wrjtMcdyj3Elfdzt9ejCo+ypcJZKXVfVRNHMUhvJVTVLr528r8rqdXmdgQOXWIQXpnrjCn7yNVaYqjqIel8dWOMpjD6iDiTpcS1hKC0UvIpiP56iZL4WIV5QkV3+LORqt4TmProEx+gs7DIcDwhiiO0NgRSEjXq1GsxGosTDhUG5DNNob3HlHEOPZsCjrgeM5yMuXfvHteuXWNzY4PTZ04ThApjPJM1SmqlD6hBCTDaYPGgXxB4WZSUCueEB06ThLzwAIkKAiazId1OF11kSKEI4wgZBTz77NP+cjmHMbb0BhUUuuDe3bu8/PIr7O7uMp5MuH7jBgeHfXIhUEHoU1VVgHAOZxyqJkBJdOGwEqwTpNaS1OseLMLhrH8tusgwNgejiZxjoR7xCz/7GR4+v0mWDcnw8phHLp3h+ptr1E+dIDQ5jdBbJczygj/45rd49Y032dvrc/PGHcAD0+1Ww4fvdBqsLK/QanXu65JVAHUYhtTrdV94SEcQeGC6KAryLC89Jgfs7u2zu7fPYf+Q6TBF64ITJ9b40Iee4dSpkzSaDYy2LCwu0W63vdl9mjIcjUBKH4JUFJ59KARJHP3IC5mTm5vcvXuXUCqcscSRT6QTzgPw/eGQIAg8s1lrZBnsFccxUgru3bsH+CKvVa8jQ7/Z1Gs1dFGQpjknT54kOhgQhBHWKuIkYZbOfAMhDEuvX3+N4yj2RV/5t3//D7/CyVOnCVRCEMSMxylLS8vcvn2TOA4rtj/CWsIonsv+vVTeM+f8/zvvWeaOH1YcRhdksylJHBElyXyxqh53tHHe7wF5HAc5Al3FEXjrKwSEcERRxLMffpbvfutbTGcTtDPzcKsgCFCxIk+zeYfZGF0Cqb7rK0vmjLMFMlA8+fiTREoxm465dOkCSEuoJEWeMxoNWVlexluJpGhrkCr0fr0S9nbuUWjN+uoKe3t7TMfjeZhYIBWBkGzd3iJQAYf7BxRFwXg4otFo0O8f0mq1kFIynU5+WFPyp3K46mDl/GZnjcY5NWflhyX7XUrrQwkkYL01jA0lZy6d975aQrB56RwFluc+/IxnxtQg6tb935GSXFmkgUBrUGCdIbAObQ0agcs17eVFPv0/+8s0Gg1SaylmBSIM0FYhlcAIhxGCsFUnjhteAWEtWVYgBd4wXRwFb9z/Wt8HVO9fK/zXD3qIvdt1Ou6DVg1tLdl0RpBb/uu/+/eQzRoXL13gn//6P8dMcyIrEFrQSNqceXiTi48+wmRnylMXHodujatvvs7ea9d5+/e+TvSJZylChWv2GE1GcHaNtxqCZtDj0b/xy9z9x7+KfOkNnv34ee5uf5EzruCN119kaWkG6TK/9Y0XWH/oNT79gWd5YX/C3/2v/j6f+os/xxtbt7lxeI/AOv4PH3+O3EF+7ZDgm9u4lTGEju7bQ+70Yt661GHxoQt879brxIernCam0+mQZwVx4OMN/AaksNiyDeG1IVo4ZMV4xGASB0IhZ4ZnP/goF/5vf4d/8t////jKl7/Kub/1y7QWN8isRVgN0pFRgAhxhFghcU5hhESIApAYKQlt4VnjUmBFSuAUnXqPJz7yV/jmq3ucXVAsLdZB54AiJMW64B3v5x91G/y4AaxCOK8uUorltVPsbp1jvHuDpBlwIlLcvbXHzRvXuH7rLdY3V/i5v9hm9dQiCB8GJbC+yHbOv3cYhNM4m4MtwBY4q8ForC78oV34hiIORuMR129c5/SZk0SJrw8CApyt5HgFwiqsC8B59csRWoVXD+EVWCVaivAFjH+O5XPyrNEylsr5g8UccMLhnMaFGlNYJrOc2UyT5wJtFI3mOt2FOkKG5EKiTQWtOXAaawuMNmSptzGI4wScRIUBSI020/JwI7BGoGR4bH68cy2oatKKTbi7u+vDWYRAa68cOHPmzDxY6P3xx48HrZnCMGRhcYGbN26QpinT6dRbW7U6LC0tsQRzuXKWpuRFjp+urgxhO2KjugpYmrOcj2rCI69I3xgvisKTYPyTmgNq1WeEwJb+pVL4ZrzAA7Hersx7YkZlPZ6W54k0TechQpPJhHq9fsRYfIDJ2O8fcPv2bZRSXLlyhcuXL7O6ujpv4v+kMOp/UsdxgPvIju8YC1h4ksdxQLVSZ/gUnj/aCsBag7HlzxsfJGWtXz9OnjzJeDxmkmXUa3WSRh3tLOOpJ1IEgUSWDX7AE02Y8/1ZXl6mKPJj09wH80mJ964OAtrt9pwMVnn1CiHBVreHnIfoelKLwGgLypuOyzJ8TWuDUra8Z/Q8YNthGEyG7B7skzlHJGqsNs5STGq0ghx99SVae9dxwxnffGmHk3/1L5ASEA6mfOnzX+Du96+SB45JFxqdFsPBiP39fRYXF8hLhWK32517pi4tLc19hm/evEW73WZ1dYX7eYI/+SMNBFFpIxIKiVKSKx94gh2Rc+v1t5gNhoikhunUWHv8SU6dP8/owvc4+MZ3mdy8RaADwqLEe0yGntxjNOgzvfUWo4U2slWnvrRAPRBcOr9B88mLHO7ts3v9FrvXbtGfzEiSBDkeEezvU7t+h0AoIicxAaSBYSZyClsgrbe+claSa0MtTGgEMfVuk97ZTZqPXkFtblLUeuQuRgpFe6Ig8L6rUgYI6ZtLgfT1RCiP+ahaz/wXUpLnBZFSLPQWiMOQwAl0KXIJw5Dl5WUcMBqP+fZ3vsPP/MzPgPS+qVI4sJLzFy5inG/aVuFQxvmWrTXWW/o50KX9hqDMe5kLbMq7cI4FgMXd52Nd7SEOMM53KI7HY4JvViDBGYs22ltuIkF4fNDnA3mSpV+XHLYEWn0zRx6L/fQff1a06j2BqSdOnmSazjDOMktTL9UQnn0qpEAXZp6iGAQ+6b7yPRmNRoCj3W4xmUzY2tpia2uLX/u1X+NTn/okJ0/7ZMUwjDzIWTLwVLlZBlF4bPPEy9HKzV8FIS7PUUoxGAwI44goCJhMRtTrdSbjIY1mgzSfkkQxEgjDYB4QJWWd1cUuTzx2BWGgyHJ2D/Y46A8Yjse8dfUqb775FuPRmFmaen/FUZ8wisitZToeoFRAo9nC5imzIsMKiJKE8bTAFJpYKmoq4OzGCf7i5z7D5YfOoM2EqZ5ibYw0Bcu1mMcvnGMwGhJKw0KrgTOCvAgIal1u3b5HOt3Daj/1QhVx+aErnDt7GrAkUUie5+zt7zMejUpvU0/DjqJonqju8PRvKaUvVkTAdDZjNJ5y994Ok9mMLM39dQ8VjWaHj338E8RxTBgFHB72WVpcwjmfpJmmKY16HRmEiFLmL6XE6gJjVOml+aMb49GIxYUFZmnqJeaFf71VKqSzDl0UDPp9n16XekuCidGk6Yw4jsmzDGf8xnf95i2arSZRGa5jjaFeazAc3mY0HiMVBNLNA9U8SJ37YDbhk/XAB7hJJXnppe/x/PM3OXvmIkoFzKYZrVaTRr3G3t4Oi86z/ayxGGNRYYAKAoIguI8S71wl8T+6vtYYjC5Ip1NaHV8EVJ2cI+CjDCGoDjzvUmhWXmzHQdrqoOb/13L5oYfodrscHh4gIwmB7/bkee5Zv0oBDqmOmOlQPmfnC+x6zfvYnT61TrNWQ+HodjvU6glFkTMqNMuLSwz6A1aW1xBCMhwOmaUZVYDLcNin3W6Tpn7z2jhxgsODQ5aXltC59gzg2QRjDHmeU6/XEULQajVI0ylFkRNHESfW137wk/EnfPzxh1q/IRntLR1y7Zl3MlBzZYISDiG0Z1VbVQa+gBOWwlVeUw7rym5iVCY3KklhCoKy7NWmwAmLLYvcvDx8y9z6MDxbpgQ3a0ycweYgUQRWkktJpgsGo0M2L51DxTWw3hzd4cPzkjias7TdsddegSPvD46Km3cBVf8oIPX4OG5XEhtImi2UkPxn/5e/w06/T+o0/6u//ctcfe0NXnjhBb70ta+z2GjyH/4v/gbfe/FFnvrE8/zGF3+fm2bM//4/+U9ZWV3j3toa1zHEaUEtihmfaJMpR9NJXCGY1ho88fyHeO1XfpWl5AbLJwbI2XU++oEAl1quXt/l4UfPocOAwhX0Islzzz7O1W99jZCYKyvrRAT8X//T/xP/9s98lvPpAkupZrK1TfPujGGesfD8OdZSONSW4cxRD55kT2+we/0NZsMX+ezPnkOoFGsCsHVwBYgC8ExG4SzSWlzZpa/HNbLBmG59DVPkLPYi/oNf/t/wzLOf5L//lV/h4M42n/zzn6HWauHmlj0aaR1aSnAh0kmc8F6ZsVMoLNKCEQIrQrQDS8F3Xn6RVDt+6d/9Jax0vhLUEuNRsgfe58or7scfkLAYZukUKxxRrcelSx/nhv0q4+w2C52AOFzm7s1DJjPNW2++xe/93m/z839llUbHhxBWh32/NpXJ4tZ6Tydj576PeV4wGI18WKq2FGlBrCLSScbO9i77O4c06x0kEUYZgkB55VVgUUoghSnxBDcPKwFKXagsS3t5dBhxpgR6zfyQ4NfhUpEy37MdWhekuWYwzslzSxA0iKIFGq0GSiUYJzB4gAAMAUesD2O0P/AbgxOCqFFHKoVwhqwoUELiPd0lFQ3WOcOc6PHAOL42ZFlWJm77BxqtcViefPIJGo3G/Pr/NEpPf5hDCMFCb4HDgwP6h32KomBvb4/RaEKtVqPVatFoNOi02gDMstQDUdOp95os2USuPHhKKFnGHMn4XeXXzzx9eR62WqUvKzUPoqrAzzCKiMszSC1JqCW1MuXcy7O11vT7fR9im2X31YpVmG+WZfczG51jMpmwvb1FlqVsbm5y+fJlzp49S5IcSZjfn0M/mvHgPVuB6RXQLsTRYyqm6YM//0cPD+xrXeBcTJzEZc6BJ8aMpmOvlCznRKvV8mDufU2dEsARR17NUoqSTOHJKdZavxYfez5KKZrNJkp5u4Hbt28zHo09G/BYU3k2S1EyII5qJElYNo08SmWNxZbZKnEce5agOVLb3t66y3RmIRSsNTs0ESiXUdy+Qefmy9S15DuvT0k+9CnkubOEMuWVr3yDuy++QS40h92Aix94gkcfvUJdthgOh+zc2yWKQ9rt9n1s3+paH//+ZOKVxZ58VVnC/WSPyHhrncBJpINCQiOu85kPPcfrJzf4zle/xminT9No5CzENGPan36G5gcu0f/2K+x/7buMt+8SqRBtwJFhzYAgHXF454AgTuDGVWpBQCeICJcXkaHh4vIqamJ4+8ZdDqaW3GU4UpJwShz4Pb3QGov3dDUW0izFoTBS4pSiphw9ZtSHBzQHd1na2WL50mO0rjzBuLeMVS0KoXzdp3ytgBM4Z+Yy9iCUxxShFoNjcHjI0uIiqdacPXOGQCn6Bwe8+eZbPP70h9DOlURAS6vd5pnnnp27UhW5RskAVSoPtPWQZn8wYHFxyYOd9ihcylsC+gbqZDqhVgZQuRIYrexlqlE18+Y2a+VZwVpXNmOOv7vVucMg8LlIRaHJC0NVNxlztFcJIX3j0HnUwFF60QtR9q/NnHB3zG/pX2m8JzA1S3NGw3HZJZpRTxKG4wlKSpJ6HW088h01aoShm5uCh0GIqfw084Kt23e4fWeL0XjCq6+9zu7eHh948gnOnz/H6soqjWaTpJZQydGDIECkklq9hpV4r1JkGRYFhcmJEm+YW2s0CZQkTadzk2VtCpT0C5tDeP9GKcu6tUzVFQKJQMmAsBayeWqTk6dP4hw8/9Hn5iDaLPV+PDdu3iLLc65evcYbb7zJdDojUo4ohKzIAUEsInq9Dqsry1w8e45eu82Tjz1GEofkOmWWFeSFA1ugnCGKYtbWVhHS4axGRjFhEBE5xel4g8cfu8IXvvglrPMAWqPZYmNzk5OnTrHYa4OAqN6kyHMm4zHpdMre7i7gJdWtVotarUYQxkRJMvcdctYv8Af9Pl/+0lf46te/Xs4xD5ZMJlNeeeVVnnjicWoqIo582uH+/j6D4ZBur0cce/BMlxKIbreLdI4g8F25H+VQKmA4HGKdl3Q3Gw3yLGM2S4kWI8bjUbmZtMrN0IcmRVFMkWZIK9hY2+DO7dvcvbtHt9lmodcjCALGkwkOx3Q6JYkSwiBklqWgvG+aFWIeLCWF37CDMPB/OwyRcUg6mfL1L3+Zk+snEWENcEynM1bWTzB6Y8hkcAimwGqNTjRRnHipvqtudrBWIESAtSCVTy51TmGMN2PX1hDHNY78+Y4ADyn94cxVEsJjbJJ3fuUp8GL+A34zlsqDkY8++gh37tzGSuv9svDS66LIvcVHVPOevEJ4T01nSGLPCu51W6yvL9HtNNhYXWDYHzDLUqJQcXiwT71eJ53NKHJFkRfUa3VqtRrTyZR2MyIvCi/3adWZDMcluJ+STVOss/T7fRYWFqjX64xGY6aTKaPxEFE2gPqHAySS8XiEq9u5lOz98e7jHWCZ80bqxhicjLBCU2t1aDYaaF0wHh6CNShniaQiEEDofbCVEyAlvqQQqJLFWlgPmAbWSze0wMsujEFjybUlKLvNxlqELjuV1m+2rsQ70AFOhdhAcjga0l1a5PzmCVwgvbrEeJaCddYHGibJfYqzOUiI+5GvXz++o1oLjg6zcLSuVGSmisEhpfQs9WMMleoAo5zECG+V01tbpre2QkVefvqxx/mrv/jv8L3XX+f/9ff+Pm+/+TadhR7fuPEyaVsxujpmlkl2goSDdkCLkNM7GbYHw5rj8ss7JG/f5fCpTd5ageZmjw//4i9w4zuvYGsPE648zm9+/wucX1c88WjIhcub9NtLXMv7JN2Qv/TwU9wYBogrZ3lptMdbX3+RTzzxKDffeo2X+hP640PquSXczxjVYLT924TtJtMQBvfu8s+/9C9pL6/zyY89xQee/jCGFFeUkiWX4srGgRAgPGqJdcWcmxMHDQI7ZDZNaXVbYDLICj72/LM8/YFH+c3f+Jf8n//Of86/+7f+No9cfpQ49uoDWRlviMKzxREo4QjwzHAhg4oWQJZn/OZv/Tbbd7f52//ev0+93kRb7ZPfhcPaAPlu876SS/2YA6oCyXQ2Kxt6EZ3lSzyUdLh+45tk2TaNZEqz1SUrYDYdcf3q23znq1/i6Q9/jKTVAxXiZOBnvPMIobUGYUvg0jqvcBiP6R/0yWazOeu+P5gQzAoM8Nqbr1Fvxii5Qq1Wx9gaTgmcFVglCVTJgsLXfAh8UKh1HlgV1oOsVAEnAkSAcSCtIVACJzTa5eS6YJYWTFOLLgIgQco6UbJCrRl7dUAJzuZGl16WFVgMaZqjCz3PLBBSzJu4HtQqsNaHuRpnOZYDcez+d8cojEdzpJKDF0WONposT3EYTBkMceHCpXmT830Q9b2PB9mpUejr8+oAnaae7DCZjDk89BkV9XqdZqNBFEUsLS+yzBLGlESAopiDm7ooKAqNNl5VRXl2mgdFOUeapRhTYK0tzwABURQSRfGcsOHzHIJ5loAxhiLPGU8mpFk6J91Ya3AOxpMJURhy5Idq5nLqKog1TWfcunWb6XTK5uYGV65c4cyZMzSbzTnJ5v3xox/32YQdAyN9LaBQ5R5ktMWa4thPSsDc93ve5bfjsMzSGc1mizAM2T/cZ31tnSTPAMt4PKTVbNBpN/16LDxI4xvBx++VozPR0efyec89FKuzk0MKSaPeZH3dK+Ju3brFZDKFstkgpKSmQpwTpa+vIY4jwiimXqtjnEE5DzR5u8MaQejtXW7dvM1gd4CJBI2gzlKtjZtMifSE+O4bNNIJb90e8Vbc44lnnsUFAS///td441svURiDXW6wfuUcH3n2I0Qosjyn1qhTa9S5u73Njes3WVxcIAiDY+zZ8qqXqkNvieAIQ4VzCWEYIed9sp/MeymwYKQoZ41/LQqoi5DLm2dZ+2yPl776TW69eRVbBMSmhksUcaPD2qc/ztkPP8vO629y/dXX2H/7GnrUpyliOiogG41xJkUIR2osrnBM9w7JAkezfoDNJauixUwYBkXO0GYcpikuVBxagxESbSXjaY6TMZYIKwOsFBQSAgy1IqUxsSwOJpw/cMy2MhYOhtQ+9Dhu8zQkXe8FbB3SCoyn++OEQUhLmhmSuE4gAwReiV2LEzKjfSCgsxQiQkYhJ05uIIMAnab+dwoQSnqbPOUJLmmeUktqWOPDpMMgwFjLQm+h3B+8Yrw8JdynHlBBwFHi+ZFyZo40COa2lNYdWYLgfJilcFChoRKvhjHG+TpKO/LCYLUtA+wFQlmfhUFp72EMUpZZQiUzVgjDHAL5AR7x3hOYOp3OmE1mc9ZpkfoURYQgK4GIpF6bXyghBEmSkM1S6rWk7N7MuHH9Ov3+gCwvCAPFzs4eX/z8l3j91dc4d/4cp06eZGGhR71RR0pBnMRY56g36mVglaPVajGbzeh0uoxnE+K4xmH/kMXeIgJLmmV0um1GozFJ0sAJSRR5GVIUeoamkc6DUVikFDjrrQW8zwlHqHl5QAwCRatZ5+GHLnL54Uu+qDYWrU3Z3ffSBlOa8npZtiSQyqfkAsZo0nxKlmXkeeGJDk6XKZiOxeVlhqOhL96N8b6rxqIUfOITz/PCd19gZ/cAqSSFLrh+4yYPPXSB/cMDNjc3Sa2ju7hEo9VmOhrT7S2UqeYKFXq7AhUFBFFIrBIARqMRYRJx6dJFGo0mw/GQb3/7OwjjJ/9hv8/LL79MURRcuXKZTreDCgKkUoRxTBQnIAQHh4fz7rfWmkatRp5lRzLxH9EYDAc0Gg329/dZWV6mXqszGo2Io4jhcEBvYYHJeMxCb4GDg0OazaZnk+Y5C90FpJTcuX2H3sIiKghROLZv36EwFisFQhiSekw6S1FSEUaCJIrQeVGyhKCW1JBKUegCpSSNZp08KyiMJghDXnv1Ne7cvsXG2QtEScRwOKTdXmf9xAZ3b9/0RWc5r5z1HwrKeeqD2pyQJYhkMaVlhbX+oCykIAzj+SLhawU7P/A46QsFIfzB8/7zz7EQqnmxUbFWytRV682sL168wOe/8HmyIvf3i6QMXpOldxEEQYixtgTnLUlc4/SpE3zkuQ8TBaCLlDgM6HZbNOprzKYZWpu5zKzVbNJsNJhOJtzdPmQwGPLw5Svs7OzSWF3F6px2s8P29jaT8YRup4uxhm6nQxIn3Lx5kyiKaDTqtNstOmXCYaAUYb3OcNBnPBrR6/Z+pPP0J2G8W1F75Knrpf3WgQgTkjBBA6NZRqAkzXItzspDUyQdgXOlSMMhg6A0LHcY51ASQgRW27ks0Gg/z4zR88JVlxut0fNEPQrtgwutExweDplONCvraxiXs3H6FHG9jrFeguJZXt5vxxhNVKvN1+d3Y12+P47Ggz6IR5JOWzY//W07HA7RWrOz4/23oyjizJkzbGycoNVqIqSaM+GO5lP588IXQE8+/BB/5z//z/jC7/4O+9MJQ+04tJannn6WxfV13rhzlVD1+UjYo/O9OzRWFzkwMzZf3Kc+dfxhYjgZnKHTPMVwY4Xf/dbLRE7BVsjm2b/JpLbDr3/tV3D7v0/RHjG2q4TTjOGbN1AWZsUu9RNLPPvBh5GhpLm0xJmnH2Gl1aHdac59l4E5aJxEMc1ms5Q0e6a2NhoplBcouaxcWxX3Mf3LxG0hBDIICGt1ptMhk3FBq9WgpmKss8T1Oj//V/8Kn/jMkF/91f8v/+gf/gOe+uDTfPzjH+fs2fMEAt+Rl45ABBW8igpinLP0Dw/48le+zNtXr/LII4/wH/z7f7u0pfGe8I6gBFXvn/v3+d7y4w6l+rmUzcZEcYJQEoMi7qzw0JVPsHfvTfZ3r9JNJoTKIlllZWGZb33t29y4eofnP/ZxFldXCKMAGSrva1pyck1RYI1Pi83TlLtb2+zt7tM/HDMYp0ynBdoKgsKSWs03XvguMpB86ANP0utY4rpDuQBLgCLAigAlNNKpikPh2RwiR4oQSVB2hwwSAxQ4m5NNR+SFJNeO6TRFiJjCgFQ14lqTpFEHIbFWgAg948SW6pI5W8tR5H6dzbKsDGLxh6fjVlFF6bMJIJHl4cNL5qp5WzWleYDZcZxh6Jzj4PCAVqtZKhI0QSA5e/YsvV73HUzC9wGx9zaOMwGN0UynU7IsRQiIIk9oKYqMLHMMBn2cc6UE2gOcXrHlveoqS7A4TnyYiRRzRVNVEx5/T8fjMc5ZkiQpLRuqEB7m3pJZlpXPKZurgyqWXJ7nczsycF6eWqqp/OP8vlI15KbTCbdv32Y0GnHq1Ck+8pHnOHPmzNxy7I+6Nu+PH9140CP9CFQNSBKJEBIpFZPJhNls5h/jBFW22YM/f/SLfW1R6ILpdEq93uDw8JDBcEiv22U8HjPNczY2Nj24I31tYo8BqKJk0h9fsx6s8zwQhD/DUx5Dyn9L4oSNjU2azRY3btxgd3fXN4+1P+vUavW5ZYW3PVNM0xlaF3Nma+XxC54Ytra6wtryEtfv3fXBvC5gpjOmWzdpT8cUQZNbiWP54Q9iBdx85TVe+saL5KMZixdOc/HZpzj98HmkCEi1RViDK30jl5aXyVtttu9uYYxmcXGRxcXFeXhb1QSrrru1nlRynEH7kzqcP456YBAIyzOsExBpwWq9x8qnP8PrG6/x8ksvs7u3h5qGtDptdGipRQHtp57ksccfIT3cYXjjNv3vv8nw5l1cobCzKZH0jdV6s8EwGzMrCib9PqoQBAREQDsCpSKUDZjhEDpnajTjwDCoQ4EPP7fGEEYxmdFEccg4lKRWMJzlHOQDtqcZF23K6SKl90RGePI8w24HG0o0BaEAaX24GsI3rWbTlFpSRwUOJwQqCtAl3jUYDcrMIkVreZFZnpUMz4KkUccVR8HSzjmazTrOgjaOIstLZYH0tWJ5T2mtEVIQlH7J1V4Q1xKO7kLKpv8RwaKCVW3Z1LPHyRrm6AasrBuLwvgAbSdJM+NrH4T3nzcG6xxRFBKoUipTgrLzEkVUSiuOlDU/oLbBewJT01I2XXnc5Hk+97mpFgkhBOksJc9z0jSl1WphdOEZfLMZw+GQ7e1tZrO0LNb8gTwzhjfefIOrV9+m3WpycnOTVrtNr9ej1+vR6rSp1+serEUy6Uz9m2VAhVAgyNMUozWz2YQkiclTjdWCpFVnNp5Ra9TmBeLxDpqz1eLp2QBWeI8JJ+/vsB3/7PDAligZiVF5aHE4gurupZw0gNY51hQUeUGhCw9GlwteYSxBKX8NowQhvWHwLE1xzl/zOK7R6cBnPvMZ/uk//WfYskjZ2dnh2rXrXLn8ELdu3+bM+Uvs7tzjxPoJnHWMhoZ6o8m9e/fodWE2myJkwNlz55ilKZPJZC6JGY0nLC8v83M/+xcYjaa8+v3XoAQybty8QaELcq356EefI809Gzg7PCQIAgaDgU+Or9fp9/torZnNZuAccZL8AKboex9LS0ssLS35/3EwHo9pt9sIIbh79y7j8YRet8vBwSFpms59msIwpNfrMRj0kVKQZymDvT0219dJagln1jcYTidEke+2J8kQay21pDnv9AQlk0NrzSydeZao8z7CBJJiPEU7R2ZTvvfi9zh55jwu8J5Be7u7rJ5Yw1rH3u4O0+kUZ40H+3HoIECGEVgfImJtgbEOqT2YIRAY7T1ykiSZH3qOFws+WL3s9szn8zEmyLxR5BcwN6fDMmdneyaLB6U2NzdpNpsUw0PfUdLlvYW/Ft6zJ6fZbJIWGb12i5Mba5w/s0mAJRCKWr3BZDhmdXmZezs7dLpdkiik2+kQSsl0OPKJrrMUJSTdVgtXFCjnsIUvuCeTMXEU8fjjjzMqLS50oTFas7ezy9mzZ8lmU2pJjTydYYxmPJzR6/W8zDtJSGezH9EM/ckdx32xqkATC4TVoc1ojMN7mTpJIAVxveGVC3nOdDrGGU0S+i5o5amGAKdLHQYCXVpjUDJOsZ5Br0vmsys3cGP83DROYgrPsmp2FmguRCwsLRHXG2ic9wN0rkx19c+9OqzVarV3MHyq++Z9QPX4uB9IPfqaOWPIN2IsCwtdpPRhhIuLi6WpvOM73/k2r7zyCoPBiPX1dU6dPMW5c2dx+CK+0WjQbDZpNpuEUcjq2gr7e31u3r3HzdGYy089w9aNW/zdv/d/Z3O9w5/72U+yZjLqypH1D6gvt4ieu8DBtV1ULUDd3GM0uccdPeLkapu3r73Gd197i+nA0FhbYaG+wKWTl3n0oSfZ3DxHt9YgbtZQcUSoArrNFiKQOAEKn2btcBSyanAf8+MrT3tVV73QzL2ynfCd+2o6vRuDp/paCA9OZNkUYyzT6YxWq+ML7TTDOVhcWuA/+o/+Q3Z3dtnavsuv/dqvcVg2MzdObHDq9Gk6nS61Wo27d+/y9ttvcvv2bXq9Hs8//zzPPfccS8vL98/7ipBz/B0X73bY/PEfzkIQSPJ8TCw1YZB4YFHFLJ+4TG/5LHu7V+kf3EJJx+qpRTpLK9y+dYPf/+rnabXa1Gp1tDHU6gn1eoN2u0OjVp/bJR30+1y9eZ2t3R0OJ0PS3LPnrZDM8pzcKtR+wGtv3qLXXeHUSUVba5LEs/aiKPYM4JK5JfC+boFTSOWwLi8BBj+f8jQnTXPSNGM4GCHiFo3WImHUI4q6CFnz7CvpMDbDCW+FIpzAlt7qWvvwiKJUT/m63REEIb1ebT4XKz/947XyfTXzu4Dt8++VBytR2oJ4JmE6Z6tXf8MYw8mTJ+n1eu8HBP2AhjEF0+mEosjKnfnowFoxgtI0ZTabzYNijbGEYTQ/B1TS7IpROLdTO/b/qkxW9mWiYzyezFmn1frn55rfQ7UuvDoriuZ2MZUNQMVaBktJcLpvbcyynMFgwK1bt5BScvLkST760Y9y5syZIzn3u4A+P8lA0E/LeHDtqAKYKkl55YdbFEXJPjb3N1gf2H+q9zTNUpQKiOOYu3e3ieMYpRStVouiKO5TxczP/OVpfP47RRXGfP88Ob7G3XfmL5+XUoper0etVqPb7bKzs+NB1WPAbAWcyvI5wHHv+KPmUhRFmKJA6IL1pWV/DYxgUBRMVYO3XYKymuYHnyVYPsFsMuXVF18mnaXEnSaPPv0U62fPEiC9LYsUxJSqK+sotCGOY86cOcPe3h5Zlr/jdYp50CFURJqiKN6hKvppGvMGlFRcuHKFk+fP89Irr/D9l1/h9p0t6vUarVaTQZIRxCHR6gbdtVOsffAZzGDIdGeXg9s3Gd6+w3B3n8PDAVaHRMaCVGhRgPRB0MZZbCFQQhGiaKsQFYRIYclERmE0VjqMsBRFipSCmVWoQBJkFh3HTJRgnGdkb9+mOJhwem/Mqacm1K48TrZQJ48oa0xHIUoliwQ9ywikb3IJKXASAudl+lESe8KWlAhjiJVf56Mo8uf8cr5WYGd1Dx2/F4TzrG1tPZYVSP93KoBSCAHinTkU/t6SeHWtQxuDsWaOL1Qe8EJ4qpco8Y2iKPy53hisLZURzvqwNxWi8T6tWmtqSUygVMlCPXoOFd/cP5H5jPiB1bXvCUw1xlCr1ajVahhjqNfraK3n/yaPFU/W+rRPZ73EXAD1ep033niD0Xjkw6sUKCURDhYXe6wsLzGZjJjNZly/fg0caO27qO1um+XlFRYXF+j2eiwsLFCr15k0RjTadby0fJFsNvMy4FabIs89xV57pmjl5Xn84wgUYG48e1/BKI6lUB61cDzIaEtWn5DzU5Kt2H+em+t9L3XhKdW59wPSuqDIc/LCBxdI5VP3AhUSRQl5rkt/zYisKMiLAlUWMI888ggvv/Ia3//+q6XP0IDvvPBdOp02YRiwdesm3V6PLM8Jo5hWp4dQAYUxDPsHdDtt4rjG9tY2zWaTIPCS+Ol0SqfdQxeGTqfLL/zlv8w/nvy33L5zZ95dvr21RRjHNFtNHnnkEVZWVlhc1NTrPigmiiLu3buH1prV1VXiOJ6nFv4oRwWmVazp6WQy39zOnTvHZOI9pKy1bG9vMxwOWVxcJM9zhqMhYRTgsHS6bQprmU4njCcjRuMho/GEKA7QukAXhjCIvH+acCRJ4m9kowmiELIUWW6kxlpyo0mC0CeexyHffelFPvbxT9JeWiJOIg7392l22iytriOkYniwS1Hk5OkUpSRhnHg2M36BMLrstlDNQz/Hs8w3MVwpxauGEMLnVwgQotToORBS3Fc0YF35GHFsLjuwxrNSS1BKOEen02FpeYm9wz1UFJYeQY4s84yDMAhBe9P2Ri2m3WrykQ8/Qy3QHOzvUq/VqCUJywuLDAcjAqkospylhR6jwZB7W9ucP3+e6XRKrRt7loyU3L5xg1a7Q17khFHI8uIit2/f5nB/DxWErK2uctjvE4UhCwsLZOmMMFDEUcBoMKDZbGKNYTgc0u122d/ff/9A9x7H8fXROi+3t8556xSnsM4glSIvNEapcm+AIEmoRwGDvV3SWYp0liBQ3v9USS/71xpVNtmUkKU3IT4R1Tmsk7iSieocOOG9cXJjyLVlaWmZZrtNrd1FG+ttA6wFUdrCCDGXGabpjI2NE/OC8d2k6++PoyHEuzF/fOEuJd4APs9RUqKiiEAFrK8tI6UiCHyw3sUL5/nzn/lzVHSBfr/P9vYWt27dYjga0et2CcOQ7e07XL91h729Q66+9jrj8Yjexjo711/jm1//OqE23J7u8cVfm/KdvI5QAbqXMIkMptAoLSiygv7+AYe7eyz0OiyuLHH6wkUe/kuf4+yJZc6eu8zy4gmclGTOd+sFzttVWm+Wb6TDSUvoBFiLFmCVQLr7g1YqFv+D4KouzH0gkp9X8A7Ucn49BT7ZQhLHNbJsgrWO0WhEvWZ8CBAeAHFWs76+xqmTp3j+Ix+d+7ZtbW2Rphntdpu11TVObm7wzNNPzWXUxpj52u/K10n5lI5Ls6qa5h2HqZ+A20KIkEDVyLMCrS1QoKSvB3KnIKyxvPk4naWLzMYHHO7dQAVw+fJjhKF/6WmeM5mmDAd7XL9xlfHIs6EWl5aJkzrTNOPm3R0maY6xEukEQVAGKRgPUGaznN17e7z00ksYk7O+vESr1SCOY6IwQgWKMIqI4tAX/kLgjCNEkeYz0iIjLTRWBoRxBxl0iOpdFjoNH6oiBDiJca5Mt/VNKRWEWCPnjafpdEZReIKDbyzLeWpuZb1xPNznwVGBX+/x6nPcT3A8Ht8HslUN7KeeeopGo/FTeVj/1zWM0WideY9xZ+eKuTiOQUiCQPrapwQys9L+yrnpHPQJgoDZzCsQK8Zn5e04t4kqAVYhJNb4QFvPaBal6s4HAldnLmM0ee7T1cPQg7gV0FU9b+e8LVgFxM5mM65efYvBYECtVuPChQtcuXKFs2fPzj0s3x8//uP42aIC4o+D9tUZdJb+yWSGOTiLYDablQxrxc2bN1lbW503BCpP1uPqmSq5e86uLkEb7oNZ3rn2PWg9Un1dq9U4e/Ysa2tr3Lhxg62tLYTgPnxBHPv5ar2lVO1WtUMURgjrkGiEM4hMU6gQs3SStNEkT6dYQmppytb1m+wfHKCxXLhykfpCB10UBBUhQXlVrShJA8YUFAXUa02ajRaDwcBLo61Dlsxdfw8eAavVulEUxU/VPfYgQC+EQJafRRjz9JNPcfn8Q3zv2y/w6qvfZ3/nANmISNotWkh0HDMKJWpxgXBtiY0nr3A2Syn29ujfuMXOm1fJt++xs3sPMx0RS0+CMk4xszAzklnuyGRI4RzCSlpK4ZRmqiDu1Wg0GjSimPF4Qv9wQO4KpJWEKiKs1bg3GmL6+wxeyVCFZjWoET1yjoNujSIIcFL4XAk0sZQYLDOdEZahVCIQhNbXuUJIRG6wob+fClsczXHrr9N4MqbZbFHJ8I2xc5JEdU09OHms6fEA5iCEOP6tEh8ErKNwR/hhVSOUica+CJMCawxFnoKDQvtgb29R6c+AAs90d0JgtUEisFqzsrwyJziII37jOwkDP9BZ9h7B1Irt5701Qmq1Gnt7e/Obrl6v02g0yIsCKQRhEJKlKVmaEscxs9mMO3fu0D8c+EOyUrjSCPrK5ctcunie8WhIksQM+n0G/QHOeZn5LMvY291hd3eHyXhCo9lgbXWV1dVVFlcWWV5ZIQ5jDtN9ut0ug/6hLwIChSucl8UXWdmxCoFgfkP5oAaJ8xF+9/koVJNEcDwFvfJ7AFx1Y5bdJnwX1hhDoS3aeK8pnXsQ1egCoz1YKqRfvKXwprlWFExGIw72D1hdXS1BQMlwOCAKQ9qtNq1Wi8999rO8/vobnlGZpezu7fHy91/l+Y9+hP7hAb1ej/29fZrtDrVag5q1nD1/ju98c4/d/V163UXvudrwRT3A4eEhUgQEQVgyYVr8wi/8Zf67f/JP6Pe9NCgvCm7cvAHAbDrjoYce4tTpU2RZ5rtwUvLdF1/kEx//OEEQMJlMGA6HpQT9RzeEFCwuLLK3t+fDour1eTd+MpmwsrLCYDBkNBoThhGdTofRaOTlDyafd+29TN2xu7fLyc1Ndnb30c4xGg9YWV4mib0cVQUh1vr3tJrnaTnnpZQ+lMsYpM6BHGEsudYMJyO+8+1v8ak//+eQyqfKbd/Z4tSZM6yvrRMHiv7BPoVxyEwjgpTQloee0LOyhfDgklI+iMSHAWSEqosz5r7rcvxdqM5hvn6oAqmA+eIjkIJ5wEV58i41L27um5UkMb1ehzAM0WVKaxIn5LJin1jCIEAFvpBx5T2wsNBi0j9kc3WF0WiMzS3tRoPx2BEGir29HdqNNhfOn2d7exuBN4ovnyGnT53i8PAQawxaC7a2tsiyjCjyr3J3Z4c49utIo1ZjcXGBN954nazdptvtkuuCQhvSfIy2kNQaDEajH86E/Ckcx72n/GF96v3R4gitIc80QngGVBSEoMDZnFBBWmhUXpAIhy0KnPJAnC1noLYVo6Wckc7NzcO9XYRXvloHw/GIWqOFkyFnL50lTrycPy2OgAElfNfeOYeQUOQ5szTlxPo6URS94zD/Poj67uOdoEcJ1MymWGu8nUmp/lDyqKSQsmKs+v+3VUKNE3S7HbrdDleuXAFKJme5ltp5SrhlNNjn1u1b3NnZ5dlHHyZNNWdOnUXnOZtnVylmKZkuqBEilcLWI3qNBr1ej7CWEFqHDkJCLQCfci+p+5RQNyEUlkDXscr6A42QFKXdgLRgpSPzWBzSHr2W+xqvpZdvFba3t7fH4WGf5eVloDTSh/LzuwNW/hLbkiEQkheqTAH2bBzrHEpK4jhBCFX6mlmMKXBOEUUhZ8+emXvECSGI5VGoUsU0tMfmeLXyW3cMSHvgHng3Fu07n/uPFygmCIjjFkZr34RRDhUIkAaH8T54UYdGr8Xq4lmydMJgcI9sdog1QxSOhWbIcrfJmc1TjEZTDg/79Ad9JqMxaVZAnhPgiFWICQKMzdCuwOCQxqeXz8Yjbt+6DnbK7vICi4uLNJrNklAQgRS0Wm06nS5OCASKJGkTRk1qtQ0vxYtb5NripFeXWGdwNkQ4SvsIU362OAfTsbfd0drXlkIIwjAiDCPi2LMDjxrdDmOO2FIPvpfv9t7f11CzRwBF9WGONa5GoxG1Ws3bu1hDHDd54onH5k3498cPbkglCEJVzgMf9JPnOYNBnyiKSeIEpcK5t36z2aTRaPi9tTzMCiGo1+sURUGapggh5l8bY44kzECeFxhtvZIgDFBKemuzcg2UUvigNem9VCs1oitZ0tVjnLMUhebmzVsMh0Nu3ryJUoLNzU2ef/55Ll68yObm5hwsey9rzY/jmvRv6njwvaiA1CoYOIoixuMxo9GINJ1iMO/4HcdBMKHAaOuzK5KENPVhVL5xkNNud+bNm6P54o73DTnag4/+/Y96vvf97XcBVS9dusSZM2cYjUbkWTG3aENUZy0/zysA9Qik9OquIoBQBwin0ElA0IV2LSKZJYzTlJHWTChYPr1Ju9EmT3M2L5xlMBlRaE0naZC0G8hAlhZuR8+7ChSsroXWet74PV7HV/WcKEEzv3f89IwH3zcoox2tQzoBTrDYbPOpj32Cj33ko9zcus13Xv4eb1x9i93ZNu1Oh3q75ZUlSmKkIBSW+sIiq8uLbDz7IYwGPRuTjYeMD3YZD/vM0pS0sFgZYQiZSoUzDj3LmaUzBrMJD3/gcT70zDPUagkhgsPdfW5du8at11/jjZe+y603rzKaZcQ4tHVk4yn5629xioxzxVP0PvABDhohWQK1AuLCIkOYBTAuUmIZEwXgtEALV8YElfFM2mcJWVnNU4mzPrCqXq9DiW354lcctza+r16cIwji6E6a1w4lecLN34eyhq6atNYgnAXrz43W2DmhwVrtCYnWlYoIg9EarQ1WeMWFlGqOM+V5jtUFS71eieUd92e9fz8Qx/77gxrvCUydTCZz+cdxn6XFxUVGIx/qMxwOfRFV6LkTo5SSJEl49dVXGQ6HFNpLOKtO98lTm5w7dxYpBKdOneTm9eucO3uWa9euUavV6XQ7OOE7ntoYrl69SlEUvP32m7z80ou02h1OnTrFY48/xurqqg94GuRoa1hbX6PfH7C4tIQKY4SQhDiUsyipvOeDEEjhE6GFYE5TFqb0PnVuLl06WnyOGCbGeA+UosjRRUGe5xQ6p8gLsiwnzzJ0kWONpqR/HAG9ziGl9QcjEZBnBoUlnYyxuqAoyokjvL9KFEWsr6/y7Ief4Rvf/BbGGMbjCbdv3WEwGLG2ssybb77JqTPnODw8JKk1UEHA4cEelx56mIO9HfZ29wnCiN3dPc6cPYMKAhYWFhiPpsRxwng8pl6r0ev1+NQnP8lv/fZvk2UZsmQc3rp1m0CFWOtotlp0ux0PHoYB50+folmrEcYx+/v7KKXodFs/0Mn6Jw0Vhrzx1puYQrOwsECz2Sw79H4cHByQ5wWtVptms0G/f0i73SYMQw4OPYt1NBpxeHjI4soKokxdDMOIIstYXl4mjCKkzKnXasyKowW6Sh6tilhrS4+WKCYQEhsGWGcJJWhn+ca3vs4zzzxN3G7R7rQ5uLfDnZvXOHXqNEsrK0RJjYODAzKtMdMpsdaE+QwdBjgrUUGIsxIXhgigyDK/COGL5SNWNRxfNIQ4dkASVeKduO/xDsoulysXy3Lrtw5nvS+JlIKlhUWM1ohAEUjFLJ0SqKCU26gyVdPfY0WREUjHZNinHodgfcplEoWkeVZKCzXKwXAwYHl5hbX1dZSU3L17lzDyDN4oiegP+7SaLTqdLrPS762SYBaFmbOPq+CwxeVlDg4OCKMYKRVJvcH+wQG5HvrgvLLb9v54b6NqNlEClqPhCK09U7269iDR2nomnXQYJXySeGHKbqEPnHICpCmZLhXLv7x3rK0M5B2zovRGtdBbXGTl1FmiuIZUPpQlM3buyRYEgS9aKBNVtWZ7517ZiFshjo+6q9Ve9D4j9Y8e9x9M/TUyxpDOSua/DMmz1AdFWo1SwTEg1VIxW4+K2XeCkkK4uQRQOM8StUISr66yvLrOB53ESYVQAQjfgRZOo0qFiKkYFjis0+TKonAEFpRUOCsppAA0oU5xUmGlQmBRKgMVIkQIxmIKzcHhHtPxmOlsxrSUFOEgkGoOFkkpccZyeHjI9vY24/EYay0nT57iAx/4IEodNW7nxvr3XYfj1+DIT81hCFSAteIYWySHMCTPc5I4QTiLLC1XhBM44T3hrPVAsECVtQv31TBSlLZEx8FgjjFTf7BT50c+LAWIArBEYeTrqLwA4YhrPq1YYzBOe3QSoNagXT9DIE5hsgnC5jidk82832O9mZE0ClZOpBR6wiwd013scfLsJnnuffzGkyGTyYwsN4yHBcJFhEGEUgLjEsLOEo2VTdrtLmGQkMQNpAxpdxao1RooGWBRuCDx/l9WYVHklT20szhtcFZjOAIsTRkMdHh4SJJ41qtUCmcLorgOJRmgmn/V4fodllcVwaBqqJbMwTkI/8DcfZB1Jkt1gXOO/f19aqUftcNgrW9snz9/jlrtfSD1hzHCMKTd7rC7u+sPneX7rAvNdDzBGL8OJ0nN21iUdjphEJZNsIo8YssgKb9H1ut1ut3ufcD6cXbd0Zw5agpZ6887fp55RtFxwH46TRkOB+zv77O1dYc89/Xi8vISH/zgkzz88MOcOXOGxcXFOenjONvwjxrvA6g/fuOPa85EUUS3251b/AwGA4bDAbPZjMIUvs58YO0Bf06vGNbNZpMsy5hMJn7vk5JWqz3/2lmLE5WS5KiBeNx25DiY+ic1Dd8NHPbew36eWmOxpgzj1cV83lcha9aa8sOfp7AOK33iuJIhDSyFApSgHkcwHjPOCqZZRtJq0FtdIYgj6kmCEmWYp/ZBgiY4AsUqxp9SGQjmFnRCgnwAUK1A38rr+KcNTIX7iRLO+fNFVW0F0gcoGxwiCDh17gzrJ0/w1PY2b1x9g1macev2HaajPhJJEoVEtRAVK4glZJKaSAjDOkGvQXvhBG3njdCMrzDJcsOosAzGQ+7t7DDe32Xt1Cl+4Rf+Oqc3N5FCoK3BCBjPJvR37nD19Zd58etf47tf/EP2r90FK7hbZIxmA+5eHWGLjA+OHUsfepK9FYGVERJfMzaCEJeljFROPVQo58HUAIUEdIl1GQBtvDqmhB6F8PZAXmtTogcPnJWOj6pGKI9bOOHeWSdI6SXd5T2AM14Ji0NYg7BVVorPLpKVrYAxhGGEsa4kXNg5GCzLdaBqHpo8o1lL6LSa74BO33lX/+D3ivcEps7KdNQqoTFNU3q9Htvb29Tr9bn/Zpan5Fk+70gLIebSnsFoRF7Sc40xqECyvr7K8tISg8EhUiq6vQUmkxlxnBBGIQkJq6vrjKcTpPB/u1FPGI9G3L55i1mmeeXlV7h67W0uXrzAhQsX6HS7dNpdtu2WZ844Qxg1vKwgDKglNeLSNN1PHOkPH1IirJgj2lUHx7Pxjg4atvI71UXZ6YEsS8mzlOlkwngyZjYaMJtOmc6mBPNOXOA7uFKVXhQOISVSBhjjGAzHNJttwlBSrzcZjUZI6dP2rOnTbndpt1p84hMf5/U3Xmc0GgNw2B/y9tVrDAcHZFnOtRu3+OjHPs6d2zdZXFyk3W6ze+8uKggZjybkxSG6ZA1sbGyAgBPrGzhn6XU7ZGmGsZrLDz/EdDrli1/6ki+KjGU2m3Hz5k0PjBc5D19+mPXVVZyS7O/tcf3aVVrtNhLY3t72/p0/wjEajqk3mgi89P5w0CcIApZXlsmyDHBliFrObOZlRa+99hrr6+vEiQ+parfbc1uLuPT+XFxaIru7zXg0otNpIyRkeUZW+BRE4YTv6JRsWOvwFg3KMwWMdhSAk2AKjcGyd7jHa6+8zKNPfwgVKOq1hN3tOwRKsnnqLJ3eAkFcY3fnHnk688xOJQmTBG0gihKcK0F5LFrnvnCwFqsL7zUiBK48TM27c7j7vu/KFqr/unLbKhsM8gg0q/zb5rI9Y1laWvL+qFrPF9CiKAjDqFz4HHmRU48UK0sLLC8uELsMZyzj0YhWu8loPAIBvV4HpTzD6s233+bazeusr68TqpAwiii0ZmVtDWMLby+wt89wPJo3cZIkYTqd0ustcO/evaPCwFpqjQabjSbTyZQ0y+g1GqysrjIej7hzZ4uVlZUf6Tz9SR9VV08KLwMMg4AsTTk8OGAUBPR6vTKQD4SS3j/IGhr1NrsHQ2pB4EE25SnSyoCUjsIYtCmTGQWkWQ5CsHbiBA2haLTa1Op1LIK80OTG4UxBxf7zDU3v7yucwBjN4eEBw+GQXq/H+toaSVn0vhuY9P549/Fuhw6tC68uCVS5l/liPAzDOevNF1TH2ZgO5+R9h5L7GRJHf8MKhxUCryRRQCljQyMxBM4gjEQrX4AZjyqirEULizIWgSWXgsSUyhPj/aSczJhMNGjB3a0bvHnjTd6+fgvhBL1Ol8Vuz7NGigLpQDpwUlA4O6+B4jim1WyytLTEww8/TLvd9s/eOfLcW654AOPIRsI9cIi5//PRh8OU7NJjAYHOBxMKC0YpUCHymJm+KNf1SlAgKr63u5+ZYx9g4VQf1vhi9n4VwwMAxnHqwTvmxo/HSLMJhcnwyg2I4wBdlKE4GqQKcPiDi5CacmYBAiMkTjQg7OACQdxwJD5CFucKtJ6RF1NAc+qSK/MBxhT5BJyX99frHQQxeWaJwqRUbtUwKvCNcxkQBJEHyu28/+nfeWvB+AO5EBpH5tn51pbqN28fkRcZSkmUEhRFhjGaJImoJzUECqVCpIxxmDkL0DdZS7m2EniHKj8bpLg/NOI467T6/wfBMzh67yu2axUq42sEU4JqHiQ7f/58OZ+472ffH3/2IQSEQcTK8gpbW1v0+4f+H8rmCQi0LsiygoODg5Ih530R4zimVvMy0zD0CjUhAyrGWvlrgGOSaeZHpPveUw+cyvK9P/LMzbKUfn/I3t4eB4eHTCcT36So11leXuahh05y4YI/vy0vL8/9UN/9tb4/b35Sx7u9dxVLtVar0W63mc2W6Pf7DAYDRqMRRZG/o+lTKeqyLEMIUQYJe+uKCiRNkoQwCHFK+ebSA2BoFIU+sKds7P5x8+pBpuof93pUoAhC//3YRccee7TfVueoQnt1nZDeogOtyZwkNxIdKITVtJRBBgFZrYFwkBYZ+UATjMcEcUSz20HagFiEYI/W6ypkKs99YFCtnvgaiCOViimTzqVU9z23nyaJfzUefN8MJUvT+a8tYEKBxaIQBEpxdvMUJ0+uk+UF/cMBw/6Qa29f4+DwkO2dHYpQUF9sI7VBmymhiolUxNSV5AEs2no16mQ6YzZOmaQzlpd7fPojz/OBRx9lodFEav9cnBQgJa1Gm/aZBhsbmzxy+VEevvgoX/nCl/nWN76OGBa0lGRvnHH9tes07o05kQ5pfuojDHuL7IcJgQhoFY6WU9xLM5yU1IOQSk2okLgAAldaXgpXhk15lZK/X5jfZ55E5RDcX7cCvIMgIOC4LdgR0Fo10wyuPJtTntVsSUpUUqKEx8acs+RFTmEMQeQB4rn/cWkvlmtP7NLakKcZRZry0NmHaDcbPuPAHe1NvsL74e4d7wlMtdYyHA4Jw5A8y0AIJpOJB45KL6Qiz7i3c5deb4HxaIIQXp67tbXF1WvXGY7Hnu5bMjxAcPbMGYajoZeaS0mUJGxtbbG6ukpR5Ny5e5dzFx6i1vTg4qVLD7G3u0MUhNTihCzTnNrcIIgCXnv9Vb7x9W/SW1ii11tgaXmRbq/DwuIi9XqLRrPJ4uIi9XptnmgZhl4aKIVEBfKoW+WOhZ2An0wlS8+ViWNa67m/z2w6ZTDos7+7y2AwIJ+NKbIMIXy6e6Ne8ymZQYCxvtjMiwLjHKPRmG63x9r6Os1mjcl4TFJ6DgUqYDQcY2PLdDpmOBwSRwHPPfssv/07v4sQkslsyje//W2uXLrAz/7czzKeTHjt+y/z1Ic+xGjYL1P8lhgOBpw8fZo7t28zHA29F+rt26ytrbFz7y697gL9A+8feePWLdbX13j8sUdRSvL5L3wBjDe2HgyHOOD27Ts0W00kgl63zdraKi+99CJPPvEkQRiiBPT39354M/ddRn84JApDms0Gewf79Dpdms0meVGQFzlBoOh0O6Vn6l163S7nL5xnMpkyHI6IophZmhIGIQcHh+SzGUu9HsPRkJWVZQqdk6czBv0+eZ6jgoQoiigKPyc8UGt8OFkZzFYUPiHZIXBSEoQRSliKNOMPvvYHnL54nk53kXqjwWw25s6tWzgk6xubtNsdwvV19nZ3mAwHGF34QyBh6W8aUSWr53k2ZyQIv1wedYaEX6Srw/L80H10svG8KGcRwh+2qnkvwHdPncO4o3AB4RxPPPY4n//iF7h+4wZhElEYjZSenRuGcclOgzxPabWagJ2/H+PxBD0a+LCUeo3BYEAYhkwmY5aWFjHOcXCw79lFJdvXOMvBoQ8JK4qC0WhCr9cjjmMODg7odDrMZlNu37nF+voJ9vcPiJOYycw3YxYWekRRyGQ8RipFq9nkzJnTNJvNH+k8/YkfwlukyBJIjaMQHUeM8pzxaMTe7i7tdpt2p0etFoPE92iDgO7SKrrIUR6px2jvpWacxeUF0/GIRr3J0vIyzWbTz1clMcY/dpYbjHFHPdMS6J/NpuiioJYkCKXY2r7LYDAgTmJOnTpFr9clDH3KuXWlG/D7QOq/4hAYXZTMs1Jq7OTcEy0Mo2OhC2J+kHiwG3z/dT/6Ny0kTnh/UmVLXrywICq7EV9DGOm7/9I6krIRWkiHFJBogRWKWQAWi3CayES8+eYdDrMRt+/cxgxHPPfMh/i5z/0CiNLnqmScBirAWYsKAgqrfSFm/RoIzH3QjoNQR6/ZF4pKibJ5fHRAqViBx3+mYq5UplPOla+VUjHj/Gvyy3hZqFuLE+aY33PJbhVHB6EH53UF1VY1TvUYrTXG2vk7cBxEO/67jjfl/qTxrwv0GI9SrBaoQJSSeItQEEjAWXThX5NPjvUsPKlKgB9HGAhwGucshbNgfUCUJEBFHWqhD7TEOOqNoKKNlpLlCmSU80OIT9d1BE6VhxIoCn9Y8PusAWswpkAgEc4D/w7fkLXWkmYp4/GUKIk9GSCKAX9QThLPADUlUcEaiy33aETp419Kv+dZARaqcEnB0Ryc//sfsy4en+9VjZNlGc65eZinMQZtCmazKY88crkMmnofBPvhDkG3u8Dm5sm5V20VGuKBd0UQGKxVZFlWNns8e80rqfwBNwxD4lrdK6pKIKqy8pLSN1C9137J+sM3eYqiIMu8wmg6nTIajeep7dZawjCk0WhSrzfY2Nhgc3OT06dPc+7cOdbW1kowNzxa3x5gAP5R33t//GSNBxsx1fAAZ0QURbRaLVZXV5nNZhz2D9nf32UymRypDIUrmXPMyVreD1r58F7nkAiM8mrBpJbMgcbjf8fPtWOKEHf/83mv44967FHdA3DE5q5CsqKIcm/RYJtYKTFCs2QzilQzGU4Z1mPy3QOMyXA4VBQRBgG1KCazhoPhgKwoqImAOInnoYZSyftqDQ+s+ucShsFcNQy+XjieG/FvQoZEpejx40jVJPFhowEC6UChqIWK2mqN9dV1Ll58GK0Nh4eHDMdjgiDg5tZ17u1tEaiY3Z19BkOvBNg/PMA6Q71RJ6nVefjxK5ze3OThc+doRBESB9KHh7lSmSedI3IKZxU2aLJ28jyf+NwCK5unWTmxwVf+2f9AsbtLTfTYkQXp7nV2/nDKRaVZ+eSnGCz00FIwUhKVhCSZQ49zXCJQceAZobKsUUrClAuktxCyZV2tvfLFYhGVRYV36/PX6V0aqse/L5xv0PomvcUcr3XnXgF2/j5IVREHhFdjl03fyqrIubLxbHz9KqXP5QCBLjQ6L3BGU6RTLpw9XSq2KiKCP6u+gwXwQxjvCUxV5c1fbZRxHPvPUcTMGNI0BRx7u3tIIWm3O/NE0DfffJOdnR1ms1lpHOtDFi5dukShC5SSnNjY8LLc0AcxLS4tc+3Gdc5dfIhZlrO2foJ6s8OdmzepN1oEOOq1Ondu3WFleYlaI6HVbFCvv8E3v/UihwcTrl27jU8Wl7RbTVZWVtg8ucnKygrdrgfZkiSmXquhggAZCF9UwxHFuOyyV8ENzlpMGbxVFJpZmpKlGYeH++zt7HJ4cMh0MibN8vJNl9STGmEQUmjN9Rs3mKYpRlvSNENIRxxFXL78MEm9QZYXtDtdhsNhKZUMOXP6DJOJL0ykVIRRzEeee5Y333yT6zdugbOMRhP2Dvu88cZbPPPhp5lOxrz1+utcvnKZg/19Ws0OjWYHYywPX36Y119/ndu3b9NuNymKjHQ2ZToes7K8QhAENC4/xHgyoVZL+PAzT+McfPGLX0QbiwFGwyF3y5TmIst45PJDdNptuu02w8Eho/GE0XDEQw8//EObuO82Nk+e5GB/HxCc2NhgOKyAesFwOJgb4wshOXnqJP1+nyiK6C305smSjZoH2qezlKihSNOMTrfn0+N0TqvVoj3VdNpNdgdT8iKbb/Jh4FnGURD7AwYglUJUh484RhiLMBoXh9zavsNbr77GB59+FpXELCwtIw4PuLe9hc4zNjY2WVhc5MT6GvtRRL/fx5oCISy6KFBlsrSprAWc8VJAfJJgBaYKKZHHQiSqZUXCEdgqjjpLVZEspAcDRMkm1+5oYxbWsby0xLMffpa7d++infWJxCpECjk/PBmjCQPJ4kKPwaBPYOpIqVhaWfYJ78Jy9frbNJstwjgCKak1agyHA7QuygAFR6/X4/qNGwgp6HTaxLFncKdpyngyQSrJwtIid+/d4/zFC0RRxDRNOXFincP+AYEKaDRqCIS3LAkCrl+7zurqylyi8/74k0cFOggpkIEkiAJCE1GzZg7GO2c4ONhnd2cXay1nTp/yTIBmg6TVQUhBIAPCMpQNeQQArfk/Agjvf+qsVwOUxah1ntHvjAfVBOCs5o1Xvs/NWzdo1Ou0Ox2a7TYbmxssryzPU2RdBaLyoG/U++NPM0QpmyvynCgOkDI4dggW9wUYzOXjcybTn8QEKf2cnECVTGNXAqlHpa/ECIEl8GyNwGGFIzIWWXa8i3JtTDQgLfeGh3z1D77OY48+yrn1dT7w2COoUGDQSBch0ChniVVIgcWIsiDElQnnmjSdzYGDitVV+b5lWcbOzg7OORYXFwnD8Mi7V3mJobXBnNlvjEEX+tg1ssyDhJz1a3nZKFBKeHl+Kd/Ls/xonRUeiPU+7FWghMP7ZR5d0/IXI8WRb6oQZfFa1jTV936SQYtIxd73VCmE8loLJ0rssCzycRAEIbkpMEJgBBirUUp6OwXrCKRCOV/jWGdAOg9ei4p1XfqXUu6nRpXzFoQw2NJ/1zqBQyGY4axBUAY0OUehLThdJrA7lArJMjD+1EAUhWXjU9JutYjiGKUCwGLLED/nJHnm540taa7O5f6QZO/3mLxvvStZJ3MfY44YqRWpoPp6vueLI0/UytoIfNBUJQv3ctKc5eVNHnroIr1ed86MnWsafkLn1o/r8B7jkiCIOH3qLEVe8NZbb1CUYaDzUJw5sCrxbP9y7SjfDh9MlTIYDdGlV10lSa6CP+BojajqXj8/jsBOvzYmxHHEameFVrPN4uIiKyurbG6e5OTJk6ytrZV+q+EDIX33y0PfHz/940FgRilFrVYjSRI6nQ4bGyeYzab0+32GwyGTyYTRcDIPS5rNZozHPjBnYWERKSWj0Yg4jkmShMlkgiq9fJMkmZOp5iehUpU3/x9+MMD9g6/rONvf/yWHpMAKi5MxEPnziSyoNxJaSYve6jprJ1P6e/sc7O4xHI+wxlsE1MKIJPHgaiIUQRT7YG8hUEEwf00VgFvdZ1Iepbcfb15Uar4ftaL0X8dQzgegW+FKv3IIbPm+423ujHMo5wtJLRxWQeEsIlJ04yWW9BKhlZxaWyWVKQhFfzAiywu00Rz0D5BK0Wy1WFpcJqrViJRCFhonDYV0GOF8gK+1hCgC62tYS4ARigJDvdfj8Sc+iMgdK1GD3//1X2Nyc8CuLDiMJPagT/zlbyGHho1PfpTp6gqDdocsCOgUMZPpmKlJaco6GocLBEr7UFVpBU6X2FepNKwsKbAOaz1rVkoJD9QS72jWl9+rCFfHSQNVPSGkzzGSIpj/jH+cPyNo5zDlXqINBCrGaFf6ogqkUqXnu/d7LYq8JAI4zp89w/JCD2xpLVaeVUsG2Q99Tr0nMLXZbDIaj31qeSkbWl5aotfrMhqNy65HxsMPP1wWXJZavXa0+M1mGFPRdP3rWlpaYnNzg3v3tjl75iyHh4eenh6GGOs4ffYcKytrWCCOYq5fu8ba2jpZOmNwsM9kPEUqSb1Ro9FsYK3j8sOXeeG738c6gXRq7mc17Pe5u7XN3s4uYRgQJTGlZYQ3/g1C6s0atZpnGgZBMA/20VqXQQZeJqVzzXQ2Iy9y0jTzYVBpSp56+bKUgtxJCm3J8pwiz8nzouz+WmxZyEdhSDMJOHv2PAuLy7Q7XTqdzvyaOyd8KmyeIqXfYKIwYjQeUW80ef75j3Lr9q9SGM9+vHXnDq1mk7W1VS5euMCbb73J66+9xsrqGocHhzQ7HYIwZG/nHsvLy96v6M4dJuMxzUaDxd4Ce3s7KBmQac3i8hI7O3cByQeefILDw0O+9e1v+86Fc2U4lWUyHjEZDXn8sUdZW11hcWGBMDzAzH04fnTDWUMYBrQ77bLYL4hrDXrdDnEc0W63SWczZmnGYNCnXq9x69Zt2u02cRzTaDRKVmWINpZACkaDAQcHB9QbDXoLC2xvbTEYDPzmJJibH1dylRAvm4jCkCzLMM5htKYWx2TGF7YhigJLoTXf/ta3uHjpMq2lRYIoYWFxmWH/gMO9e0hnsTqns7jM0vIySb3BpH/AZOy7/XmeA75jo4sch/byW8ScPSBKYKry0vFzyy9wSvrC1/usAKJMtBZ4OznnQwOErTx/jvnqCX9wunThAkpKsqLACQiDcmM+FrbS6XRot710a2//gMXFBcADzbv7ezRbrfKae9o+zqGkYnV1lbt37yKEZGtrC10UtDptrHUkScwszZhlKZ1uF2MMN27e5MSJE2it6ff7LCx0ybKUei0BBOlsRrvZBGuwhWVtZYnDg31msz85TfT9cXwID1BISRAGRDaas57nzGchSGcZs8mU1155FYel3mqxtrlBo9nCWqjFCc1GE2OLUiHg56EsWdTGWm8SL0s/IZvPgaAsTXnjjTfAWmpxSP/ggLXlFTq9LiurqywuL9NoNZGBmgMCFQsL3mej/mnGgwcLV/oXqSAgKlN5ozCcd5ePMzAe+E33/U54J7DiWSX+Q5RidSPwBa/zgQEOiRWS0BqssAjniBxE2iKBQgq0BCv8YyWKIIj57Gc/i4wkAoWbKkaTEZkbkQ12GAyH7OzvsL29zcGht8Kp1+usLC9xavMkvW6POEnQRpNlGaos6rT2YEMQBGxsbMwPJIBPUb2PlXvkneqcw4QGbfTc+7IKsPQhMVWokCZPC9LpjGazSb1WIwxi3xywDoMpG1e+oHG8u0et3wP8NRb4QLcqfbvaG5SU8/v3GPY1/xnxwIHzjxrC//E/9jE/rCGFRUpfqwUymHvJ+nnlk1/BIgJDEoTeoieHWEZYYymMZjwb4XC04hbWCZJaHV0Y3wgow54EAiErYFGWKg5ZAuE5Qtqy5lWAQgQ5HryCbJZjtCMME4QTHByMUUrS7gSoAJKwhjEeXHUOamECkhLg0l4yZytw09uc+EaRmXsDVlf/3Q88Rwf7+9jOx+7ZB1nJFWvx+PfGY281VYFqRVEQJzHNVpO1tZVS8XGcFfI+QPbDG/7a1mp1Lly4iBSC119/DV2+N97731uwBEFQLhcW6yShq4AliVQh2ng/YWct1nm/aWM92/v/396fP1ly3Fee6MfdY4+75VJZWSsAYgcJbhJFamm1Rt0a6+mxfvb+iGf2/ruZZ7N0j6mtNdYa0y5R7AZJQSBAALXnfrfYw93fDx4RmVXIKhRIAASpe2CFvHkzbtxYPCLcj5/vOf2pdM3DG6oRlHL93yAISZKE6XTK7u4u165d48aNW1zZvcLW1jaj0XgIqfrEHlwgUTfP5385uGzC56KFSG+rM51OBwJ1tVyzWq07G6cVZVmR5zl5XjCbTplNZ8Pkjy99rDWEYcxsNuvEExfuS71i5CmVM59Gqj5LbfvkMv1rIQRWCBrhejmKxvlkC4mRPq1xvudSCII05lp0g2v7+2RFweLsjGZdUOQFFomyTjHer9taC13QdT+R7SZQenFYf/9esVgsuyyRmslkzBtvvOEquP5F3KpFX88z9G16rbIRLpipFsL1Rw142qKsBau7ACuDQBEoibQBUiqS2Q5WOhsdfeM6Unbe9QisFQgNVimMNBhP0Mfvekhs49qksdr56HY6J2sswTjl7d/9XdLpFulsm7//T3/KT9/771gTsyo1Z4/m+H/xd7T5mqt/9K9IXnqZIpoghYfyPIqmQuQFQRxi+sBKIVAYpJUDd2Xt+YSZkxW6En9tNPopk10XiVXRiQ7plvE97/xYK9tNRndCcesqXt0/93urXSWPts6izw+ciBMkni+p6nrob5RV4Y5VqxmPUr77nW/jKw9lz8/nl/kUeT7P1LIkjmPnCZkmPHr0gMAPsNaZi/cd8qZphrCkwPc5Ojpivc7Js4KmbjFGgm1J0zH7V69ioTtQMJ/P2draIggCDo+O2L9xE+l5TEYpSinO5md4ShEFATIIsL7HeDphNHUq2CBJaBcZnqcwpnFiJ+E6r9IqvvbSy/xP//5/RAi4sreHtoaPPvqYpq5ZzOd4ns98vuDunUeYLpEvyzKyLKMsSzylnGm7dSb+FovyA6rG0BoGEiiNY7SFuqooC6fYDUOPUeKTJDE7W1uMx2NmkylxFHH95g1efuVVojhxvrNVRRiGnJ2dUVUF1p6rA6qqAq05PjpgNhnx1puv89/f+SlCWJq65fDwmL/8i7/hP/yH/5md7V3+6d2fcHXvCuv1ijDw2ZpNieOQg4ePmM1mHD56NPjMVPUjrly5wo3rN/AbTegH7GxvURQ5dz7+gN/6ztssl3P++b333WyCtZyezmmbFs+/R6sb3v7m22RNw3T7CkVtGc92vriWexlMxbW9LZqqJQ4ilJLUumZdOuXttd0rKCvwfJ/1es1yuWB7e8buzg7vvPMON2/e5ODggOl0ymg0wfMV0p+yXhedMljihSnXbk4Jf/QBkpy6cZ40Eke8CiEZJSnaWtbLpbv5+CGysfhS0rY1tScxUkHg8/HBfX76zg/5/g9+nyaK8XzFbDYhFJJsnnG/ekBZtmxf3WE0CUmSG4TzJWdnc6dIrWqnR2orNC1VW+IJg+/cyB+78ZluoNQaNwDz/ZBel+LsKy2qI1KVFV3SHp0K0HalpXR9Dk2rK5I4IIp8GtPSdgEX2rQYq/E8r7tmfbZnY6yuSEcRRV0wms0wSOK4papy8jxnd/cKk3TK2XLOeDLh3t27TKcTgiBkb09RliUHh4fkuZskScYJN27e4O7de2RZxvb2dkfyW6bTKU1Tk5cFCA+JIIxiThZLkjjGk5K9vX2m0y0ePHjw5bbT3wBIC866wsMPBAiFle6f8HyE56O8HM9X+KFHkRfkqxU/e+fHaK2ZbW+TjtJOoe8mOnr/7b5b289WCiE4PTvm+PjIDfmEwA8CgjBkMpmQTkZcv3XDkU1pQtwpD4ayVW0GsneDXwy9x5EjhBxR5HsBUvp4KgQ8rBGuHMjy2ID5cbXj4++dr5vhfcD57XbqdmFtV5XcuVsKi7St81QVAr+7iWnPo+3WpYRA4Z7LAsskdfYuH7z7Ae+++y5SKrJsTVVV+L7PzZs3uXnzNm+88dZQctpv25NlTY93Ht3+9inZ7jMXjxsX9plBuQUdwWH77zEXyFQ9hHw2VUFb11RVwdnJgjAM2d7e4dq1fbzAdeDdRMR5V64v7704Ie9+Spqm5eTkZLBVmUwmKN9D9UQ45+eoXxt9B7r//dPayqcu8cWh1hqLGEI/vc4rTMiu7J7+fEp6sxA84RQq1hL4EdOuVF0pH2EMZV1SVfVAPAoBVeU69dPpFGParqS/H0zQBVE2FEWJtZYkcWofR2qFhIkrzZdKcO361XNPUikGDzFjnP2CI8kfDwd7UsnXE/bG2EF1elGZdL5cryzv1yY6VeHjbf3ia0eYyW4iSgxhV23bDsFTTdNwNj/h6v4O3/nO2xd898TwvRt8UTi/F4EgSRJefe114jThpz/9CScnJ9A2eIHv+ntSorRGKQ9PNbReg1JeN665MNnTKeBM3xZw7WRQvylFEISEYUCSpGxtbXF17ypXr17j6tWrbG9vk6YjfP+id6S49PWTuGySbYNff1x2zi97/j/xqY6QUnhK4amAOErZ27uKtS+7bIa6HsbqdVVhtBkCUcMwJIyjC/ekc6X9xfW7e+PnvsvP2C9cmjxBJ2ZxXIg0LqncCuM829tubrOzJhulqdteY1z44Xrt/N27YGtHKucURUFZli7QqwvILoqcunYiMK21u2avXuXWrRu89NJLxHH8xRyArxgsrpoJHu/R9C2x98p/rDfTkaKIrl/pCZru78Yq7AXxgLUgu3EnUrgcB9n1pYzFN5LACELjSPWuCdIKgZJuwtezFmGFWzcQjiJefPt1TpuMN1LN6M/H3P2zvyE1lkLCo3JF8c7fYdan3P7DP2b8+uucTVLKSUCc+1RlCVaTBD5N7OPho1FIzwWvGqeQ6VSdzmpLCIsUhr4KsOcUzg/cea9QCIHq+1pd3kovfmixXa6KxnWQGgwWYw01hgqBaBVBLTF1SxlKVBAgBXjCVdBUtav4NlJQljVNa9GtIU18Xn/tFXa2dlzu0XCt2S91/va5yNTVauXk8WlCXZVEoesYCiEoywJrNUmSorUeyn0ODg748MMPOTs7o2kbd8O0bhB27do+nq/Qbct0MqVtW/b29lBKka1zAuWhjOH44QO8/X1+/vMPKZZLXnrxRY6PjrBaU60zJjM3AH90cMD2zg4ff3QPpEAbg+ySJK0xICU3bl7n5s0bGKMZT2c0bcvXvz7G8xS6C80pioI8L5xKrm05PHhEWZZk2ZqbN29ycnzs/AFbRxqnozGNFvzov/+Yv/n7v3d+VV3n81ydItnf3+Nf/cHvkiYxSkmu7l1FCekUskqxt3eVqq4RXRkrxqn5qqp2IVadKtVZJ2jGgYcQkj/4/d/jgw9+zmqdo3zFyekZSir+8Uc/4lvf/Aa3bt7kJz/5MW+//W1OT46RnsRYQ5qm3Lp1C2ssH7z/M4SQpGnCfD6nqiq+9rVXuH//PlIJdna22d7aoqoa/uTf/luKouLDjz/uOvywWC5BWMqiIIoS3nzzTa5fn3YkmH95g/qCUFcNTeCUQcvlijRJ0bggEaEkZ/M588WC8WzK9rYLKhqPx5zN50OS9Ne+9jWWyyVVXbNY5sguEOzB6SPyPCdKE6IkdaWWnk8kzqXqSrnExKJyQQyj8RhrLY3prCOsdd68nqTRGt3WVE3ND3/0j9y8dZsbL76MVArjRSRbIX5SMz8+5uH9O1R1zs7eDnE6Y7o9I0hCFyhWFFR5QWtASh+aLslRulTnvryjv17djJoeCHHpOR9JK7uHhBQI4x4APZzXCSCcmhWraZoKYVqqsnAOrd2EQlvXSOnakzEG21ZsbbvZ4Olo0oW0ZfzsvffY27tKnrsJkNEoRgg4PjtFeZK8yInjGN/3aZqaLMuRUrGzs0NV1bRa88prr/DOOz8ZvNqWy+Xgt3R4eOgsBgKnOKq7Ge2maVh0s1tZljObTbl9+/aX2k5/3dE/VN0P8di/XgHj+z51FFGXJVVcEccFdeepVlc1ZVWR5YUr4zdwenrsUki7ELXhoW2dN6tSknGnYA5Cp34ZjcdMp1Om06kr3fIv+B/aTwb+wGZQ9ovCDT76Qbvz2/rTP/1Tjo4Ou1TdMTdv3mR7e5skcROD4+78hGF4brNwgaR5sqT8Irl6GYF58fVlXk2Xned+qfV6zccff8zW1hbf+973iKKIOI4H30c4V+tdXNfTXn/asXoeXLxuQHWDPffZ0WjktqvrxxjtDPmLomC9WvNP//QedV2zf22f7a1tojAaSmaH49NNwhqtmc/n3Lt3nzB0pZPb29uEYTiU/gkhehvNfuPO9+e59ubCRz/j8p8n+gFzP7nfk4W+7w/7eW6n0Le183NrTPsY4d1XKQVBMLQ5N+EtBt98cKKA3m9XdrYlYRiQpPEQOtqfm55090JvUFpdvAYuU3X3x7Rvqxef64+REbiBR7+OcwXqedn+xesQHleV9D+FEEMCdL9/ZenCYfM8J44jmqbB8z3W2YrZbMIf//H/QPwYaXFhq/5FKJ2+KhCEYcgLL7zIdDrl8PCABw8ecHx8TLbOnP+81jRtg+4q73r7ETj30O1V11IqlKcGhWAcp4xGKUkaO2HIbMZstsVkPCGK3D3VlVG7zz+2Zb9gQ9iU/f/m4mnndnjWP/FEOb9X0t2n3L0qjuPhHucqms4nqNxEwLNI20/eo563zX3WtjncZy/8//Ef5790Ar7HjsHQZxAul2Y8Hrtrtlukr37pJ2bPfdnPfdh7sUF/rf4mhk49E/2x5Yn+iniytZ1/oJ+c7InD84/IT1Ti9GMXwI1hLlb99LhwXm23QeLCtGm/fXAe8uj7Pm+++SYf332f1/7wdwm8gPf+y1/SFhWR1ZC3HL33EVT/D/u5JvnumxQiolYhYZKSrTPa2hALjzLQeEKAozqRQqEE3TXUHRmX6uu2UDzRRjqy2RrHNWAtuiNQRWvQEiolaB05gDIWaxq3LiuotcVISWVaGtsiCo3KIfYCAuVRixYhLdrqwde/1pqyrp1lqLFgNLdu3uZrL74wjDGMtYNSeziQXwKeu8xfCkFTlVRlOfimnp6eYoxh1pXZ9hdtFMU8OrjLep2T5TlYkFIM/jxX966QxDGL+YKtrSn379/nxo0bfPzRR6Rpiqck87MTPM/jvfkZeVbgKUFd5mBasAZPwGQyJY5Ttra20a1hNB6hpKRpNca4kitwQRBB4JPnOb6v8H0PY0F5AViDaQ1hFBIlEVvbs0GJOp4kHB8fAVe4srPLyy+/xHK5dNtSlfhhRDqesbWzw8f37vLg3n2s1a6M0DpvPk953Lh+g1dffZXAd6WBgedKIlXnsdm0rZu1EMLJnYUz7g2C0Hl1apfu67wjDHVdYowlSVK+8+1v8ef/z1/Stppc5ywWC374w38kTSJu3NynqioePrzPrZsvsFou2drephQFQRBw7fo1Vqslq9WKsqxpWwNW8NHHH/PSCy8yn885ODhgNpsBAt9T/I9/8m/43/6P/4MHDw8QUmGMYT5fYlrNRx/dIQxjiqLoCK2DL7DpfhI3rt/gwYOHhEFIGMecLebOnDuO2Nna5vDgEGM018dj8jwH4PT0lCiKXLp7loEQZHmOtYIg8MjznLYxHTHszunPfvYz937rUsq1dj6s/cWslBugF1WB7/koP3BWEU1D27QoT2Gwgxfpz+98zI/+8YdM0hHTvWsY5YEETwq2dmfMj49ZHB/RlBXJLGM62yKdTgmiiLZpyZZrsuUS22qk9ZzXFRoDyN4zi86LsO+IGAvCKREQwqnGpcTVbwtMV2PQK1yUUq7032iapsS2NXUDVV5QFoV7MHczSMYYqrJASkHoSa5e2SUI3DHRRjOZjMlVxXy54OatW5ycHhNGEWVZsVwueeHF2yxXy6FDIKWkrqtO5eBzenpKozXvvfczlFLcunWLxWJBmqYsFguOjo4AhvfiJB3U80E3qF4ul8yXC1qtiZPkS22nvym4SGydl/upoYOmw4gqjKjjmiRJhoCKpmlo+xJpa9BWf8JPqid5giDoLE6c/1XYkWBRFLnfw3AgUXs8SX5tCNRfDue+Rs4Xt7f8+JM/+RMWiwX37ztl+PHJMf/r//q/8NN/+if2r15lNpu5cxXH7F+9ygsvvMitWy9w8+ZNwjAcCK6egO879OfeXk8SPw6fpm65+PtFcvT1119HKReQ1xMHF8vyzwccjxO4T67zyfefV0V12eeeJIr75YZrSnpdh9sSxynTyRby2rn/mbUu+KduGteJfErq+vb2dqfkOSdL+u+yvTTxQq/zafv+q9Wdfjp6svIieWktQ6rxxYnFfvlzXDznciBd+/X19yjVhXxEUdQRowbPiwZloNuGtiPqLdq0ONv9x0sxL2ujl/lHXlzmyW1/cgKi//tFO4ke/T5cXP7iBMJFct+FZzaD/5mUktOzU/au7NHbHdR1iecn7O3tcuPmjS4E5nM7lRt8RlxUKAN4ymNn+wrbW9u89OLXWK8zFosF8/mcPHekqnsONy4odSBZFFIqwiAgjCKSNCWKnMovjmOCwCcMA5TnJoD6ob+7ti6o/TbY4JfEs9XL0FfL2E5VqlRHQUm6sU03aYZbtp84e16F9FcB5/vpYDuOoCfsLC7Ep99X56vNpc+Lbg2f6GtcfP3k536TcUHD+OnLXmwzXN7Pu6wv+FhfUggeuzc+sc7HX5yv0xqD3/EKs/GEb739Xd754d/zzX/7rwn8kB//lz+nKVvQDQkVzYfvs2xKXmorrnzjdVZ7u+CFxNMp89Wacp6RjlN8X4HR+ELgt60LC1bS2Wf17UuAsQJhH++vXtjJJ/ok3bExLqhaGde/dAGvXR6AtphWggbRaBpbICMPMw0xRiBNjRSGFldBZLDUbTuIosqiAG146YXbvP61r+ErhRLDVd6FTn257fe5yNQ8y/A8RxaVZYlSivV6xe7ulcE/rK7rbmbIUtcNh4eHrFYrN4BuG7R2J2dnZ4sru7u0TUOeZ4xGyRDstFqtGI/HzBdzphJkLZltbTFfLhFKYiQUdeWS+iSEoevMep6PMZa2NZ2HmaY3WPcDRRQ5jxRrDUL6lGWJFa6zoKQLJVjnmSNUgxCv8wwRgkFV05rOH0uCFRahJGVZsbXjc+Pmdb773e/w8MHDbkDWDdA6RUQYhmTrNVs3r9PUDaYrsUnHU5bL5TAB0FQ1fhiyvbVFXVWcnp7idZ50pjVUlLTaG4KUmrrkO9/+Fh9++CH3Do5ojWG+WOB5in/4hx9y6/Z/4JVXXuHhw0dcvbqPFGJQAS8WC5bLJbdfeIGf/vSnaGMwdUMYRpycnJGmKeN0xHx+hhCC0A9YLuZ4vs+f/Js/5n//j/+Jk7N5PznBKiu4//ARTdPSNDVN03Bl78st8y+KkslkSlXXlHWNH4SowKeuKj6+c5fZZEoaR87aYbF4TL0xHo85Oj6i0S1RHOPmjCyBDQHBeDKhbWrm6xV5kaNNi+f5tFWL7wc0TY0xDAoWi0Wh0NYgrelCyAxCuOCRVjdOkTkeo+uKH//kHV7Yv8Zr6Qh/PAVPuWCXKGJvf5/VYkW2yqjKkrZwpSzJeEIUJ8RXdplNp+jWBfVo3dDqelDLmF6B42Y1XCqrUo6074OphEVY56ElhPu76pL8lNenEzc0TYGucpqqomkNbVMRhhFlWSCExOoWTyqsNXhCsDUdsbs7o8jXeGlMEIYsl0tms11GoxGHh4+QUnB6dspk7LxrhZCd8tvNpG1vb7O/v8/R0RF37t7HGMMrr73KRx99RBjGXUmmYjQaOfVwZ6AeRU7BeHp6iu/7ZFk2pIXOZjNW6zXJKGWVZV9qO/1NRE9+SSkHQtUYN0nVh/a03eC8H6Rr7SYj+tLCi2qw3ivrXOka4nv+cD9WSiE9dWnn5TJF4wa/OOq67gii8/NmTTvcX7a2trAWRmXJd7/7Xfb29njnnXd45513aHVL26WDgsXzoqGdjNKUK1eusL+/z7Vr1xDChcPFsVM89cROT7pa67xa+0qZa9eu8fLLLzMajUjT9DFVYo+eGOqJr4vkqRDngTpwuZ/u09SxF3FZG/s0wrVf5uL7lzVVS1+afU6S6q4KoPe+9vyg8zl73HPwIiENdMTYJ7/j/LsvlvM+7br59OvpVzkI8zqPrv7c9hOcfUl6URQkSYIxPXl5HpTmPtcrnh4/BsNz8sLrvi255R5vM+47mwskpfcYifrkuen7Ik+So0/ew55sq3A++fAknhwo959/kki9SDD3ZHRVVQRByNnZHM/3mE5Hrk8sLa126pIsX3P9xj5vDEGjX/4AZoOnXW9u0O7GMYI49oiihJ2dK4C9EDrnlEW9as1NFHhY0z3Tu0qP80H/RQVfr2CVF77z8u3bPIc3+Lxx8Z7Wt8nhnicuEGX28QnSX3eScNhXLjx3hn2yn1zuwv53f7n0+fKbcnx+ETwvqdqTqP1yj6uM+5U8e2LfMQuftiWX/GZdgChC8PIrbzA/PuPgwX2+/id/hBaSj//qr2nOjsiUwZgcHnzMyZ81XM9yJt/6Ou3NmyzShGA7xcxLsrMVKkyQiaCJFLXnhIbCttD12S0C0wqk8rBdX8lYV4UrhLMv6Pex7wc1wtAot4e+FtC6iu3KGoxpaduaptTYWhKZAF9LxqMJVWgoZYtuSqIGVBBSaU3dtFS1pm5dPtByuca0Ddeu7PKdb36T6SjtjlU3kSJ6da/rjwwVV19ws34uMlUI50lS1zVt2zAapTRNhedJwIWPCOjIipD5YsnZ2YKz+bwzQBeuatgYdna2kdL5+vRkZRiGWGuZTFzq4+l8QRC7Ms6bN2+yLAquXLlCVRRMd3a4e+cOViksgqp2nQI/DMmyHGsZzGulp5hMRoyShCRNQDqyKy9KDJKyaoiCgKLI0NaQ5SVpGmO0JvB9tLZUZUWrW/av7qONptGa2A9p8gJtHBnlKcVbb7zFf/2//7xT5HUXT6fIbZqGqq4xxrpymdARdMYYkjSl1YZ0NCYvCkfcas3J6SlFnmM7awQsqK7sSuv6QrmX4nd/8H3+t//0f7kUOaPJ8pyHBwd88MHP+c63v0XTaB48eMC16zcoi2IIDRmNRhweHbGzs8PBwQF1XeP5jpjO85wwCNje3ibPM6x2CfJFkRGGEf/DH/4r/vN/+TPmSxdCgIXVOsNai+dJ5yOG/nxb66cgzwuU75MVJcrzqJuGxXLBztbM2U9IQRCFnM3nTCYTpzL2PPKiACkYT52vXhpFjOIxDw8e0LYtYRizNZuxWC7Y9n2yvHXkvW4HMtbz/PObppQY3bowNa1dSITQ+L5CEpAXOaKT0xdNSVWUNEXJ3/7D3xBPR9x86RWS8QQhlPM08RWT7YA8zymXS4rFirZuKNc5cZqSjkaEUYgKPVCCQPhIHaOfKPVwg6fz8hiJ8y9xEn+X8ujUUD0hphDCgm6p65KmLmmqnCJbuuvCCFToY7vkQ6NbwiB0Sl1reOHWDd7+xmvcuH6V6ShCtw0YF7x1fHzU2UoUKE8ynU7I8pw0HXHn7h2nHu9I7qqqOD4+pm1bZ8OwWlEULpDFJSBXWGspioLZbMaHH37I1atXGY/HRFHErVu3ePfdd9nd3XVWHkUxhE4dHx+ztbX1pbbT3yRcNsPdK7fcvaAjVbty4/OE4G4QZw0WM5S79GSsK9/qS5G6iQUhEVIOXpo9LlMMbgZwnx88z01Mim7yxWhNWTn/reXSBRicnBxzenrK0dERJycnpGnK9evXz0tKm6YjzN1kk1LKEaCeoigKTk5OhoCUnhyKoojRaMT29jZXr17lypUrTCaTQWHYT1Q+Wab2ZHu4SGA9SYo9qQLsv//i5y6u65fBZyFUz8k8N9M//KH7KaUaqCtrQfLJ7TSmnywTw0efJM4eW77r6D+tp98T4p/WK/1VDsh6+4a6roc2Ya157J7UPxPd710YY2exIKVE63NlsFOn9m2nVzZ1kRW2P7YSnLHpUAgoOtLJGI0A585qP9mmniQ8L/u9fw2fPG+XnceeIH1SgXzRIqDf5/59a50dUn+/Pjw85IUXXnD3YOHCPEF3qdqn7O/v8a//9R8yHo8vCZrb4KuEvr93kXQKgsvKes/HLVaI57jUJb+iy3yDDS7te16EPV/QKeyeuLf+upKGl02uPfn7s1Sml1UzPLnef6n4LC3iqermJyYwe5yrU7nQV2Ag/p919IfnvHHBlMqTvPyNt3m0nHPS1Lz5R3/A3mzGj//jf0Is54RNi98UZMcPuf9Xf8nuYkHyvW8Tv3wTPxlD7NN4HllR0a5rrPURkY+PzwhJjHBCFyFd7pARaFkP/QNtm6603237kBcgnQdq3TrrgLKx2FLT1pratLRWozGIQBJ6Pr5QgI8nDGmRYc4OeXjvHuXODpMXXqKqGupGU9ctVdWQZRlnp2d87fZtfu8HP2A6SlEDOW37OcRzstt++rH9vPBcZGrv1dAPjuu6Zmd7m7qqqKqa5XLlFCRBAAj++Z/fG3wnEcIl7GqXePvaq68ymYxZrVbcunWLsshpmoaDgwM8z2M+n+N7PtZYbt64he8FfO2lr5HECfcf3EdIRRwnWGvwA1caWBQNTeGMloWUmG7SNQh9Nwjb2emUKQZjXYo4VgzKKEcsKaq2oW2dPNl6zu/H8wKXmteFbijlYbTF90PqLEMiMNoSRxFXdne5e/cetu80c95xL0pXru/77pA7w3dN3TRdKnaA1s4boqoqjNbUdY2UgrpuMK0mimKEELStRik5NODbt2/x3W99i7/6679FCEmeF0RhyD/8wz9ydW+P69ev89FHH3Hw6CGj8Zizs1Nu3LhJnMTESczMbDGZTPjZz96naVpGoxFHR0cUWc61a9cYjcYsF2dO9di0PDi6x/6NG3z/d77HX/7lX5PlJWAwRrBar3nwyPnUrLP1F9Bkn47VKuPK1auMJ5J1tiaJY9IkYTRKXALkyg380zimqiqSJGG9XndGyG5Q4QbqiqIsXEp1EDq1qXJlfQ8PHtE0Ncp3MzVVXXaDrvOSdIQgDALolMBKCJq6ptYGX7nEU2Na6qbGAlEaY5qG9z76OdMfTQmUYP/aTZLpDkJ6GCFBGNLJiDhw3r5ZllFkOXGckEUBySjBi3z8OCAIQnwCpFJ4gYexruOsrTN8NtaR+sL27iwWJdx8mafOXVysaWnrBl0WZNmauimpq5xsvcAKQTyaIGqJ7GwLBgJNCPzQY3ua8ubrrxCFkroqaWqnvI7CkPF4hvQ8ijInzxw5Gvg+TdOws7PjJhFazfHxMdPplCAI2NndZT5fEicx73/wQadwc0ptKeWgWpxMJkgpWS4zjHE2JZPJBGut8wnsyLrd3V3Ozs4oy/JLbae/ibisU3dRbfUkgfAkAdB/xv3ry7x7z7bHOydP7aR8ynsb/GJQnYVCT2BaazvbhpI0HXHlyh5FkQ82Dj156mCfmKg/J3KCICBNU9I0ZTQaEUXRoEK9WOoPTyeNnlTePastPFn2/OTyl5GwF7/rs+Bpy3/aetzfXffP7TNupv3C4MgpBs4Vta6nKAfS5HxdT6772cfp07bN2k+u86uGIAio6/oTtiFu3zVKuTa1Xq8ZjUa0bTu0o7IsiaLIDR6kpPcIPlfMi64NnYcqCSTWdiFOuMGOlBJjNdhOgyIkVp8PXC+W2/ft+6JP5ZM+p277n1JidwEXlaWDdyAX76vn6+iDo/rv9X2fk5MTrly5MijHjTGugkYoqqokyzJ2dnb43ve+x7Vr17pU7F+tEnmDx3HZuXju82PPlT2iH4Wer+Wyb/ultmuDDT4PPJXUesoyz1ru1wHPs+2DYvI57ge/zsfiV4HnPV7PXO6SPz1jOmBYnxTODziygp3ZFt/+7m/z93/5F5TWcvv73yUIPX76f/5HFg8rsAZsgzg94uwf/h6KFf78DaYvvYzY3uM0DPHiBFN5VIsczhqE75GFHkWoCDwPBQgZYIULjzf2vF/VGQcigLZxfYkW5/NPWWGblrJpsHgEXkjqh9hAYQNBqyoCXRPYJVIr7OkK8/6HtO+9T7Fesf///vdkdUtVaXStaaqGYp1zdnLC9av7/Mkf/zGzyciV9huDlC7I6xO9oy+LSeUzkKlZltHUzvtOdDLxxWKBFAqBUxFZC6enZywWCxarFW2vius6d1eu7LK9vUUcO1/Ntm05OTnFaMtqvWI8HnF6esru7hWOHh4SB86jZ51l7Gxvszw5Q2DRVcN4PEICZZnjeR6L0zOati+vd53S0WjCrZs3uX7tGmWeIYVCCklVNySjCQACg7WG0WhK0NkGKE/RNq1ri8ZQ1w1lUeL7AUr5GN26EJxWOyVsq4nCiPFoDFwYsADGQpblrJZrGu38Mtu6BhyJKoSgyHOU8gjCECElVReoo5SiaWpkl8IrRDWU/dV13XkT+mAtX//6W7z77nucns0RQjJfrhBS8Dd/9/f8+3/379jb2+f+vXvUgbNlOD09Yba1xW43MBZWYK3go48+6kgMl9qa5zk7OztEUUiRr0mTFGMMy7MzrmzN+O63v8Xf/f0PKesahMQYy3KdYR8+dP6vXyL2ru5Rtw3L1YokifE9H60bmrp2fnLK+aEmceyUl77PaDLuSibtuZ9j29JWmiB2BPdqteb0ZE4UBWTZmiRNiaOQvNADod22ziMtDEPKssAYSV1VhFHUKTUBIRBKEkdOZYqQ+IEria+qCuEp3vnpTwilRBjLrGoY7V7FC90ATwiQcUgceJArylVGtl5QZZAtJH7oEaQRURzjq4ggCPCDAOV5rp8spPM/AYRQeEK62aVBNeOIVmucmrYuS+qqpspyiiLH2pZW14AgDAM8z8MPDEI6zx6pFGhDGHh855tv8Ud/9Ltk2RyFT7ZaIhEkqUt9rWvD/Tt3mW1P0cbgex4IgZKKo8NDRKeSDaNoUNAURUld10RJTJIkFEXBeDwiSVLu3LnjlN5JMpT6+76PMZqDw0P68tW86EnygKqqqOv6AvGzwecJcUHd0pdeAU883Nzj+FkdPHc7/fwUght8Njw5WHGdOvfMj6J4UFBeJOsuI37cei6E4V1QtPVKwCfxaSTp04j5Z+GzLP95DzQuU4o85Zv7Tzy2/GVKkqFre9k67cWXzyCJL673sq25OOPf9a++qkOw3d1d7t27NxD/SqmBYBRC0DTOP7Un/YMgGNKOkyShrsthUtWFTYXdJHb7GEkJ3TEVEtWTj11pv5Tu2eqFrg+BPbeXuLiOJ1VCl53Di+Ftl10P/fv9/rlKGe/SdfSksdaaMAw5PDxkOp0OfRdwfq/WGqzVtG2NUgFaC27cuMGLL77Yeeg7bAbiv0F45qm89K7wBW3IBht8nniy7W7a7QZfQVxspk9pov18l9QtgVV8be8G/ne+x9/87V9xVBdMfudt3n7pCj/6X/5/1D+/w9VVxTaKJluz/m8/ZnYyp7h/gHzpJZIXbqFnEwgSkskI3QrWTUtWN9i2QQkJrXVkqpR4shqqVocJX8TA7wmgNQJpLBEQ+h7JJMV4HgiJ0Aav1fiZJtQFrI4pDh9xeOcB+eEpaVZRHJ8hbt3E39pmLRRNXVIXFav5gpOTE15/4zV+//d+j9l0jLTaEbqyC27thAdPHsQva8T4XGSqMQYpJNPJtJvJNszncwI/oKpcZ9RinSrx4SOWyxVFWV3wSnO4dePG0GltmoazszOquiYKQuqqQU4UWpdk2ZpxGjM/O+HgYeVKz9uGIs8wWlMVOdNxSts6/1G/S7pfLhbQqQMCz+f2jZtc2b3iiJcsw/c8fD+grJuh5NlTLmig6nzh2rrBCCiKHE8K1usVy/kZ01GKpxTW8/Gk4uzslLauybI11gqkhDSNHdmru6QxXCf2bL4gz3OKoiRNU5rGqQ3S2JViF2XJer1mN4qQQlCVpUt4BTzliN08z1guV10QkqBtXeJaXTdIIZECvvudb/Nf//wvaLWhrRuW6zU//+AjfvLTf+b1115jNBpx787H3Lp9m8Vi7vziOh/cpmpcSeXWFqenxxjtCL4sy6iqiv39fcIwZrleE4chSiqKPGd/b5dvvv11/ts7P+6OqxtUr1c59/Wjz7u9PhNNU3Nl7wpC9deVpSoKrO9hgaqsKIoCL+jKST2Fr5w/53K5HJRXcRRxvD4jSiOMcWrGpm6xwI0bN/jg4weEUcR8eTbcWJRyyYhlWVKW3QAMp0jFOAVoGEW0xmKlIOjCdPoBmjYGYS1lXfMP/+0dEIpvvG1p0Uy2dxxp4WTVCCmJk5QwCGirijrLyNcZsoCgyCgDH095XTCPIoxipKfOg3o64qLFqf8QIIwzhsYY2ralrkra3nO1855D+CBBCH9Qd/rCKVytdWnTUmuSIOHmjeusVguqcoXEBWXtbO8Mg9LT01NmsxnGaJI4oSoLzk7nSKkwpiUdpVhrGY/HFHlOURRI5TEej1ksVwRB1CUMV0ip2N/fp65r5vM5dV07X2QLq9WayXSLJElYLpduQsKK7r5lqKrmsdLfDT4/iP751tWxDKUXT5Kqwg7v9cTphjP96uAyNedFGwdH0tDd+8Unlr1YHv6YurIjuuDppMyzSM/nIUQv2/Yvm5B/lkrkeT77ZMnec6/rQlVZ96HLF3kO0dkwPWwfu1B/OSXcF4TeLqqfsO99fpMkoSxL4jhC69JZ+qQpWZYRxzFhGFLX9ZADoLvqoN42wBjjJvfXa5Ik6cjYlijyqKsGz/MHP3tnF3Aeytp7VPYE70Vl6tOUqP357r1ZL4ZTXVRq968vVo/112f//VmWDaFZ67WrGIqiiDRNzyeEtfNyF0JQNxV5kVHXJbPZhNdff50oij4xAbLBbyIuO68bQmqDX2ds2u8G53gO3vJzW8dzLXehEGDQjjxtYQHGd5XP0ghu3b5NS83f/u1fs85aghv7/N7/9//D4Z/9FWd/+hecLdd4CpRpKX5+h2J+hjo8IPn4Y8KbV5Gvv8J6PKKcjtF4zHRI2FrapqbxAKdPpTECayxt6+wgPdllVsgu3loKAhWCElhlMLalsS1C16jWEJY1/tma5Qd3WH58l2J+QtRWXIkCjrIV752doK7f4rf+5/+JapSyPJ5TrXLW8wVNVfKD3/ltfvD97xGGPtY0KF+CtRgru+NnEVZ84lj1b33RV/xzkakSxWQ6xvO8QaXqCESPIHCdwKZtqKqKxWLBOlsPJc9au85sHEfs7uwSBAEnJyckScLh4SFJkrggmCQhCEJ0F0QymoxQyuO0bfADD4QlCH2E8MmLkDAJncpTt840FyjLCuFi/ACBpzzmZ3OKPGMyHiGEwPMUSexUi0VZoKTXlZG6Ji8FziC3rpCBT1tXNE1NkbukdxXFWO3SxKwxLBdzomREUWQkcYSUTuXcnzkLVHXNyekpR4fH7F+9hpQtdd2wWq2wOJWpkIr1en1eJlk3KCkpy9ypKJVHVRVI2QewSNqmQSnReZ1ZXnzhNh/dvsv7H3yIwVJWDYt1xo9+9COSJOHmtT1Wi1OOjg65eu06Z2enbO3suoGDdGrb2WxGtl5xcHCAsZYojiirisPDQxcSIhXL9YrZeIKMY6q25ZWvvUhZV/zTu+9RdN6w1hrWX3Kwz9HREV7gs15nKE+RxgnjdMT21pSP794BLFeu7JKXBU1VM5lMKMuSIAhcUmkUU5Ul2WrdqY5dcnWe50jpSNej4/Wg7oBzxQcwqFp9zwPriHCjNUpKGt06T+Eopu0mJDx5rsTy/IC2acCPqNqGH/3kpyAtb731JqYtmEx2CKIU348QUqIF4Hl4ApQn8eMAdEtT1xR5jm4qhACtDUEY4imPJE0IgsARlta48DUpHhvcOWJXO7V1FBEEPtZTWCFodAtSopTEGgMayjxHl5ULv/KU812VAr8jqsPxiLpySp8kSVisV/h+gO8HYA11VSElXciQ50K8Oh/Eoih48PABoyQlDCPabl+q07PHCNB+kNy2zqLCKeacIrhuWpRSLJfLLijHUlUVZenKFtM4oSiLL62N/kvCM7uv9uKL5yFP+4KSZ3yf/SSZt8Evjye9xp5875zgeRbBIh5b/snXTypaLyNl+79f/P15t/+ydTzPZ35ZXLbPn/U7LiOwHjtmlzOfw7vDMe3+NBzDiws9ZdsfV0AyqFKfttxX4fqTUrK/v8+jR484PDwkjmP8zkLG933gvHTfeatKZPcsVEoNHv7OLz0kz3PyPCdJEubz+UB6OoLUeUDnRcZ4NBkm+CaTCev1iiRJsJjBC9oFpnoDidp7u/bb14dmXSRWn0am9+ruPhjW2Uj5wzOuqiq01oxGo4FYTtOUJEm6Pvzjate2df2XPM/QumFnZ4vvfOebSKnoQ4Y299jfdDxjBP+psJe8/vTn9gYbfOG42AQ3E/UbXMCzeMtnfebzWOZyfLI2/cl1GTQai5Auf+fGC7d4Y37CRz+/SzVvIEl45ff+kOXeTe7+5V9Q3PmIulyh24b4bIluG8TJHP/oEcXhI6IrV9m6fot2NEXECfge+IpWCFoBVipa4zg0Ky7YE1l7TgJbi280qjXYtsG3LV5R4C1WFHcfkt1/wPzkFJ3nqLOMaDxlntfcXc5ZxoLJD36Hr33/j+HKVeZnJ+TLJbrWaN3y3W9/i9/9/vcIAomxLcoTGNt2JKo3VGdJLF16Yjey/PLwXGTq6dkpnu9Rr5YEQUBRloSh+ymEcN5nozEPHx1RlBVV03REVO9RZtmazQjDgKZxnqDb29sDqZrlGdPplKquGI1GCCkwCKIgIE5jPM9DSNw2VBXj6YR0NIbBeFawXjvPNosz+2+15vRsTpbnTMcjdrZ3ugCttjPaD7C6BaNI4xSUR1mUSCUwdefBYA0WgZKO5DHGInVLVZUIKajrlizPGE0mKCm4/cIt0h8lLJaORHSDFUd2npzOOT2ds1ws0dol3rd1jfR8ojimquohbby3GsA6f1djLbOtmfO6VIowdAOCksL5wBpDnCTUVcu3v/0tHh0csFxnmFZTlhWPHh7y05/8EztbM159/S3e/ed3OTw8ZDwasVosGI3GhH6IUk6t8fLLLyOl5MHDhxRF6c5RUXDv/gO2trdJRxPWRYFtNUk6Ii8LXnn5Zeqm5d33fobW7pzrL1nxJz1JXuSsVgv8wKV/C2s4mc/xwxBtdZd6LFlVFUFVUzctQRAShpGzexiPqOsKz1dI4drgfLHEolitVrRtQ5RMCKMQsS6x5jxkpx+8KOlsGWyXeqeEs2YIggBPqU4Ob9HCONm8VM4CQBukkrRGsygKfvjjn1BUJW++9hp2vyVNRiSTHYIodl1U4coMrVT4YYTQGk8JTCvRvuxCX2rquqK2JVVZuMEwjihW0rqBpPIRSqKUh+cFBFGIF/ouMEvgJkQsNMZghcGTAmta6qrh6PAhuq2wxmK1IVSSK7sz/ECAtPhBgBCQhokjTj2P1XqN8iW6bcFa1quMVrdsb+8QxwlFvmY8GqGkZDwek2c5ddNQVTUnpyco5cov03REGEZDOJ7neRwcHFJVFTdv3kRrSxRGWKsRGFarBdPpCCkmvPvuP3fHUDhv5w1+hfj8Blubwf4Xh+HY2vM004vH+2Lp8pNqzItlzBd9G59c/jKy9uJ6vvB9u4DP6zu/iPV8QnF7cbkL7/VOG4gLHXTx/J3MT9htCKeIEE+QJF+da+683Y1GLrxwvV4PAVT9hCd0AaBJTFnpTmXq1KeOQC1IkgRwwarT6XSYdOw9V/ty+qpy9jPrTJDnBaN0xHK5GPzv8yJnlI6QUpLnOb7vU5blEL4Kj3umep73mA1G/50X/ad7i4KeFHaqVcnx8ckQ7ub2VSOE8xPvq2B65SyA1k23XWsODx+RpjFvv/02W1uzy4TMX6HzvMFXG5t2ssFXCE8q1jbY4CsGe8mrT0A4bkV1OSgIV9WiZMA33v42iT/m7/7hh+QadOCT/NZbfOe1FzDvfcDP/+5vOX73Z+wsK0ariiJ/SHF2gPfgAV6UEl3ZRycxZnvK6Pp11GiCSKa0UYr2fYTf9fmEAulhPQ/bCRF9ITB1Q3u2wKtylo8+ZnX6CHO6oL13SnRS4jeaVue0ieTBsqSoS4KXX+Tat9/k1VdeQk22yWqP+wdHLPIFUgGi5d/88b/ijVdfxZcCISxSuH6QFKqzPXBV5q4S0v7KHj3PRabGSURR5i59vOsEGuvCbKRQCKloWs3R8QnL1Zq6aRxjLUSnnpTcvHGNKA4oy5LJZMJqterKoBt8z0cpyTpb4QceSTpiNB7jSeWIzFajGxcUsF6tmUynCCEpy7LzFXUBT8YYR8Ra05XPVzR1QxxEWAtl55HYNjWFxflWdmSvCvwuZdoMP+tOYur5AcrzhxRhayxBEGJt17kUkKYJ48kYPwixdu3IJfqBjGS1zDg7W3Dnzh22t7ewRlMVBcgKcN6qZVkO39GXT0bCqe5kpwgsy4KmqV0ZuVTUZYUAqsoFI8y2prz66iv86L/998G2ICtyPvzwI3av7PKDH3yfa9dv8v77/8x0MiEIfFbLJePJFISgbhp005COR4zWLvHelVJL6qalrGtu377NaDLl4OEjVNC4kuu25Y3XXqNpWj786GMXrCW+3JTX0WTEar0kiiPqumW1XiGAKHI+pmVeEMcJxmiuXt1jvc5pW03TavKywGrN9vYWaZpQ1g15tmacpjCdcbZwispsnWFtQRLHVPUhAld6p7WztHABYz3J3yWTd+dTeR5Ga4SxGK2RUnRl7QaJQAnQukZ6EoNP3rS88+57FEXBt9uWq1d2KduGdDwlimNQ6twixLjWJgTOw9RTKCyx36tcLLpt0U0NaKQwXVKxwg9D/DAEIfF8H9UpTJu2pu3uTYYuvVg4NWlb5pRlzccffUBWFFh8dKPRaPb2dhhPXFK31oa2NSyrFUkcIXwPP/DxlWK9dpMHeZ4T+M46Yp1npHFMkReD5UVRFgRhSChCtLbDANba7jqtK6LIhYqNRimj8YjlatWlfEs3Gxc737s8W1PXNaNRQqsNFunk6Bt8/njWYRXPs9Bn/LrNQP8LwSeI0Wf87cnXT1vuaefqMq/VXwZPkrmfBV91Yv6xbRssNdy48UlCtfvEuebhWYrUp37fhV+eOEWXWRD8avyNzzdyPJ4QRREffvjzgdgMgqAjI9tBTV3XFX7iD5OiTdMO9gBlJxiQUpJl2eAtWlUV4/EYay1lWbKzM6YsK6bTCb7vs1pmRJGb6PM935GqXYXLer0eyvF7ErVfZ5qm5Hk+/PR9t119OGmvbNVak6Ypq9WKIAhYrVZEUTyszwUyRpTlYrDmEsKpT401FGXBeDJmvjgjikJu3HyVt77+/xr8xL/K7X6Dryo2bWaDryI27XKDx/HLtojn+fxn+o5LFxZP/7NQODq16/gh8b2QF197hRrDX//t3xLFERWGLPKY/vY3eO0br6LvPuTh3/0jD372HpyeEtcN45Mlws6pjo8JRjH+ZMTDd/6RSvkQpODFqCBAKo0REqs8jPII4gRtW5oix5QFVBV6uYZlRlA3mKqiRlBbSWkkbeChZwlXXnuRF19/k60XX4DdHQrfY54VVMuM1WJFVtZ4ccRLL9zgt7/1TZI4xhMStHFKVCHd8Rh400/4Wf1i5+CXxHORqTs7OyyXy6GESErZeRJaPM91vI6Ojjk+OyUvC4w2YJ0y0VhDGidMJhN6+XKapiwWC0ZpSl1VBL4H1ikCPM9jlKZIIVgul07N2nkg9qVRcRcgBO7AXvTF6k1xAVfqZC1b07HzIBWiI18b+kRWY4zz0aw9ZFey3XadacARYb47TLLziBBCEMeuZDpJEtcp1q7ceDRKOTw86dQbdOSsoaorjo6PmU7HRHGE7ymkkmjjysOV59M2jdsWzxuCD1zIlPv+vpx52M+OxAOg1Xiej6cUb7z+Gqenp3x85y5aa6qqYr6c8/7777O1NeNrL73I1b191usl29vbVFVJe2ZI0hFhGLIo1oPHWFVVg5qh7TzEHjx4wP7eHpPJZNjeIHAE++uvvkrTtNy5d4+237YvCaN0hBROtSgCidGG49MTdne3Ma1mMpkMQWlN7WwpJtOp8/GtKpq6Zjod4ylFHKdU+ZrFYkmj+wGKIIljtIUkjro2IYdz3Ks/+nOitXvdAJ7yYEjZVXieQikPYw0IS9s2VEXZTfyo7v6oqJqKn/7ze5R5wTe/8XWu7muauiROE6I4IQgjsBIEGKsxWKywIDQIjbUaKT08KUA48lZKAVajW4EUHspTjui1FiEFFkurzdDOlOe5SQfd0NYlbZVRZBlHpwve+9nP0dbti68EUrSssowwjFguzvA8ye72NjvTGSfHx5wtl1hj0L7PaJTiKd9ZBgjB8ckJXuAzm06oq4rReIS1lp2dHU5OT9m7cpVllnN4eERRlPi+Io5DgkARRY4kFwLiOGGxWLC1NePO3TtIqTrbEUGSJERRxNncKbKXq/UmgGqDDT4DfhmfzOdZ7ldB5FxUzD4tCOiL+M5flHD8xPZ1vz65us+yF59lWdt14m3nU9WrVYUQj/3tVw3f93nppZdYr9e8//77g7LT932iKGK5XNHULSY6L6kPAh/T6uHzfdl82zrLmNVqNfSriqIYJr+rqiJJ0k756Sq2elVpX4rfBx8qpcjzfPBLzbIMpRRZlg3l//3z92L5v6s2cX1WrZ0Fke6CXtu2xvNk1wdpsLZFCIM2NdZqqrrA2Jajo0cEgY/vS37wg++7yrI0fcyyaIMNNthggw02+PWB73m8+eabeL7PBx/8nDsf32N0fYelsKRpyPitV7j21qvsn55R/PxjVu99SH7nDu3ZHL8oUWcFclXQyhbrgbCGCAlaUwmNQaARGKFohMI2GtoWD4GwFkyNVB5V7SHjKXYywbu5x/Tl6ySvvUx07RoqHaGV4lFe05YVxemKuqopypJWt8y2t/nG229z6/o+UVchbLFdWPZXF89Fpq5XzrC+bdthZr8vR+89qA6Pj8iyjLqqsZbhn7GWra0ZfhBQFAVh6COFpCpyRuMxq/WCyXQP3/dJ0xQLxHGMbjXr9ZpR2hn6XyAOPaUujBosbatpm3bomFqc31VRFnjSlcn35v3gyvXbtnUMd+9/pTXecLIEnucjpXDeqID0PYIwoNWubMqtzwchB0J3lI7Y39/no4/ufuIYWgunZ2dk2R55njMZjwkiV94P4HnuONZdR7svyer9POuyIo5jRqMR6/V62B/f9ymKHCnAWoPvB2xtzXj11Vc4PjkZfFilUhweHfP++x8ghOD6tT3KquDBgweEYUiWFURRRBh4JEnakeViIM5PT08HH7Esyzg8OmR/b4/RaMR8PndeoUFIFIW8+cZrjnR99PAXaJK/OB48eDSc4yiKsNKVeQvhSDQhbKcQKbqwiQowtE1FFIZMxmPXNkzDeDzGtDWrxRlZUaORxFHM1lbAOq8JfN9NAqCG8ruqqh4bIHue79pZV1JoG0d09uSr57Uoz3ODJKDRGqkkshuA9W1NSMWH9+5TNg1vvZZx+9YN6iqlDAOSdEwQRPh+6JRDwpXuWxRCtFhjOwWqm8gYyj7pVdWWVreItjn307PuOpLChVI0aEzb0JYl5XpBvlpycnzC+3cecO/BI1oR0Vh3Q1XKcnh4zMNHBySRTzyKUFKyWq3IsjVVUbK7u0NVVggLRZ4hu7L90WjEZDbl/r0HJHGENRDFEXmeU9cNRVlR5M5D+ObNW0DLzs429+7dI8sylssl06krT9ze3kIINxFUlhVRHLOYn+H7AYvFkrpuKMuaIAxJ0tGX2k432ODXCZ8g7lxJxqcv97zrewqeRexcto7Pgwj6okr9n/f7fvkVci5W6AnWC3+67NvOJ/Q/4b7qPn/ZcRW4STvOSVOLHSaz++fcrx6uj7S1tcU3vvENTk9Pefjw4RAupbVmOp0OYgEl5TB5H+AmyxeLOUmSMpmMWa/XhGFIEPjUdU0YOg9yZzOjaJrKPVeCgHW2oiwrxqMx62zVTYobpHJEa29X0wdfAQNJ2ytee4Kz73P3y/W2BVEUkaQxy+WSJI0pqwJtGoxtaHVNqyusbcnyFcvlnP39K/z+7/9uZwWQIKUbAnyVPG832GCDDTbYYIPnQy8CkFIiheD1115j/9p1Fvl/5s6Pf8a1/aske7ssA02hPJJ4RvJbe4Tf+y2iZU5zfMb6/iOKBwcs7t6hOD5GNjlK16yrAikslhorFBoXOCXxECLBBgHa8zFhgDfySG9eIXrpFqMbNxjt7OElI+okZmEt86qlWbXoqsA2hqZouurpDC9QvPLaK3zj62+yPdvCWo3qq0aNPa8E/4ri+TxTT06IogjP9/DjeJiJd51lyfHJCQ8fHpBnxblPqmDwlNrb20MKgTEaJQOybOlCbnQ7kIL9bL2UkrquWSyWHDx6BFevIABfeayXK7zAHwg9hCBNU9brJfPFvFOcXvBcQzjSMvARgiEAwCkM6DwtlfMhjaLhc1LJodS/JzWVUl3aqXHkTBR0hKzbhj7tdXd3mzAMujAsN1DR2iCUpMhzsizH83zqphnS4933NERR7JQHrXYksDFo3aKUxBjn9RoEngudMq3r/CtBGPjoTr3gWY1AcO3aVa7t7/Gz99cYoKobzhYLHj56RBxHjNKY/f1rfJjn1HVNmqaURc5oNCZNEpqmxtrlEMBQFAVl5wla1zXLpaapKmazGdZaiqLA90vCOGY8GvH6qy/TNl+uF+Xp8RlJkjCbzYjCkHTkfMrGo6QjldthEANuFkcJSZqkw7lWSpKVBXpVoSQEQYSVAY12pYG61iwWS7C689U9L/GXSjr1J2CM7nzOBBg7EKtSOF/XgUStK0QQEsUhSk6oqhqL7SwrJMbY7nwb7h8cka9XnJ2d8MLtG+xsz2jrEt8LCKPYKVo7Ba0vneeuJ73BcgMLxrpEPmsNAlf+ZzToFkBgBGBaEC7grG4atGid/+464/T4iJPjYz6+c4/37zxgWQOhRfjOD1Zrw9l8wfxsSXrjClVVU3kl1hhmsy2KsqKtG2x3Tc1mM+4/eIhSiq2dHVbrFUhBOh5zfHLKZDLqlD6Kqq7wfJ+tKKYoSwLf70ob3Y12NtsayNMwDCmKgvFojFIeZ2dngOD4+ISzszOiOGa+mDMeTwj88EttpxtssMHnj8/TIuDz7rQ9L0n7uWz/49X/57y37ePePrnos/b3Mx+LwUvgy8fl1gwSMERRxLVr17h69Sqr1Yp333136LsYY4ZwrfV6hRROEbFeryiKgiAIXIm8cb7rrvwfxuMxi8USpRTT6ZTFYuH6V6Mxy+WSyWSMtWYIOa2beqj2qUqnZAUnVOgrj/rALGvtYIkVxzFFUTCZTDorApdL0OqaqhJo3bJeL6kqZwOldctyuegUsj7f/e538H3/sf7PRXyVBykbbLDBBhtssMGzIYTL6AEYj1P+/b/7t7x7/Sp3P/qYw4cPsFKxvb2DTRLytgFfkKqA4MYt4tsvkBjDbl4iVwVimVHPl1TrNU1V0lY50lcgQBuDUB7+KMWfTgm2ZpDEqDTFpCFZYCmFpaw0tjC0JzVV02LrhjYvWLY16yzHWsHW7jYv3X6VN19/lau7W0SehzIGKyTakRZd30w+VcjxVcBzkangOntxEg+z5L2y01jDw4cPOT09pW56VarbeWst09GEUepUgUIIwigkW687RaUrEa/rGmstWeZ8psqiYL1eD8Eyoe+jtWa9WrG1u8NqtaIsS5I0pSgKsiwjy/JzItcyEJVKKaSQj6kHXcmv1ykLwo7NV66BCIHolhXCKWtdWVWO6Uqmq7qmbRuEcD6nYRTRNppWG2aTCaM0ZX626FSy7ngYLK2BxXJFVdUY45Gt10wmztfLGk1dl2jtjm3ckda6bamsUypeLDWr63roAHu+jxCCoiiG30dpwo0b1/n4zl2qtsUYF1jw6PCQNEnY3t4iiiJu3brN/fv3HLEtoCpzkjRllCRorbtyap/ZbMZyuRzKzPK8oiwK8rIgTVLizh+srCpG4zFbW1u89uorX0ijfRqm06lLylWOnK/rGoFltXIDIiHEYD0RqBBpDKenp0gpuXbtGgcHBx257tHU2nnHpin52YKm1bRNQ75aolvbkfMexrjZoCAIqJq6s2xQKOvK6YUQGK0fC1zpQyfatsX0amtP4fme8xg12oUz4cakvh9gtcYazcl8xeqdn/Lo0SNeeuEGN69fYzqZEEWxK8fvnPM86aM8MRCzXufF6y5QjRAWJSxCOim/bpwSWTeV81c1tgtsa2jbgjwrOZsvuf/gIXfvPeDhwRFZrfGSEVbX+L5EoLBWUDeaj+/e49bNfaJOdRNHEV7gs3/1Km3T0nQedFEcObuL1lB2QR5RFGGspdEtxkI6SsmKgtV63Vl1eCwWS+IkpPz4LtvbO2RZTtPULJfrbvDrlLpZ7u4VRZEzHo0GdX1dN8M/Y7IvtZ1usMFvGr6s0vhn4TIi8mJa+fPgWfvwNPXe563I/VXgi/CH/VXs78VQs34b3GuJe/xZpFRMp1u8/vobVFXJvXv30G1LFIYcHx/jeR6T8YyiLGjbltlsRtl58nuex2KxGJ4jp6cnQ5/3+Ph4CEI8O6sxBpIkpq7rzmImoqpqkiQmDCMX7KgbqtpZZlV1RasbEBatG6q6ZrF0IaJauxyCsirwfWcJ5fne4P/ats6iaL1eIaXgypUrfP3rX6dtW6IoHo7LRQjxuC3EV7l9brDBBhtssMEGl+OiRZW1FmVhO0345u/+Fq98/20OHh7wj3/zD3z08A5+EJCEIbN0hAx8SiUxngRfIRKPYDTCvzojsC8wsQHCgvEEQoKxFSiDFS4jRQMGOq5K0NYNzarFtiBaKJuWVZWzLlfEvuLdf/oJk+vXeOGFF3nj9Td48fYLhL5H6Cs8DMqAtKCFRXaz/5fXTH218Fxkau/b6SnnC9V7QPl+QJblHBwcUlYVbau7jqt1akwBk/EYKR2hlCQxdISm7/ssFgu2t7Yo8oJ05NJHw8D5WJ2enqKkK3dXXZl0rx64WBJVFAVlWdI0DU3bdiE57m9lVy4VRwEXw5Ccb6TslJ9OIau0dmRqN/gyxuJ5Lnm9rmuOjo6I49il12dZp2qVA/GZpimeksymE7ZnM+7ff+AUcx2RZo1FW8PZ2Rn3Hzxkd2ebqizRrWZ7Z3vo/Cul0J0NQZqmCOlCoeIoxvN81mtHOBtjybJ8UBl4naq3KApkVRHFCdvbM2azCY8Ojp3SsHWhTHfv3yeOI8qy4OWXXmRra4uDRw+RcUxttFPJxgmB71PkxaAg9jyPwPep6hqL6LZT07QNXquIo5imbcnWa6I4Zjabfa6N9dMwm03J8oy67glzQRxHg+JXeR55XuD5Pm3niVZVLgBstVoN5XUu3GFKmTdkeY7WpiPANXXdIJSHMZbA9ymrdvC07W0XzhN3zeCZJjhXsCqpBg9fKbpUe2PQF3xwB9Wy1rSNdjYOSLT0KZqGn9+9x9HJMXfv3+fWtWtc2dlhMhnjB95wvSrlBpBCgGkl1prh+hTC4inPlWR27zniHxrtQjjqpqUqS7LVGUfHJ9w/OOHew0NOFmtaA14YIl1LANOiG4nvB2ijOTtbMl8sicMZV3a2uyCzkslohLEG013HR8fHVFWJbi1JmzBORzRty8nJiQvRiJwX7+npCdeuXWO9zjr/ZEFZ1EwmEdYK2lZTFNVg9VGWJQDL5RIhhGv7qxVKKZIkYblaY60Lr1oul19qO91ggw2+HDwZhvS0oKznIRM/K2H8LG/ZL8qb8plk8LDQF/LVXyk87Tz3BKIQsLW1DVj29/epypL3338fcH2w1XrFOluRJAlt27JcLtna2iLsCNfZbIbv+xwdHbG7u0td12RZye7uLmdnZ6RpSlVVHBw8AiAIQk5P3WR3HMcsFmcIIYkiF8Dohx5RFJBlLZPJiNVqRRi6YNYg8IiTqKveEtR1NZCxokv3LcuSJEn4/d//vQvtVHaTis9SHX/uh36DDTbYYIMNNviS8HgY6XlfVkuFLwSpkrxy+yVeufYii7M5P/7JT/jHf3qHg/UZUeiT+D6jMCCNQpTvkSmJ8EKE9ZFacufDuzx8eJfvfvc7jCcJta6xvkBjkShUC7ZqMbWzBM2LiqJqWa4LtAGpJG+9/TrXXtzjzR98nZdvf41xkOCj8JAYDJbOFqrbFWFxPqnisR39Eo7mL4bnVKY6YrFtG5qmwfM8R/RYODo6YrFY0NTNQMr0dV5B4LOzPXO+UkrgeYrFYoGUkjzPHWlpLU3dEIQRbePCefK8oGkaxqMRVVlhfZcgDkDncdp7STVNM3iq9ga1tiMjy7JAt+2gfHUfF4M6we8Ur3XdEiYuoEd3Pl9e4CM6Yi3PczBOqViWFXmeYYwm8H2apmUZLwh8FzYQRyE7Oztd4mozEGvGaISU5GXJw4cuACDwfcqyZrVaDcoBKeWQBO/IpIiyqlBCEccJ63VG07T4ftAp7kqnPggjpPDQuhxCqpI45ub168znC4qywSDRdcvZYs69+/dp2orAV1zdu8IoTSjynDAM0XWDNi7Qyvc9ZAGmbRA4pUXbtjTdBdvbHtTd+R+lI4qyYL1ekX7JXpSrtRvwFEVO1Sk4elJSKr/z8AUsFEWJNi1KCDzPle2VVUmW57TaYIzEk8IpHjW0xlCXFWVVI4SmyMvh3Dat86S9aCNhtHZBbFj33ZxfG32ScA9rLAbjiFrh2qFSPliBbh0BKpRCtzXC9xG+h25gVTdkd+5z/+Eh29MxV6/sMp2OSNOENI5JkhA/8PCVOk93tiClU6Qq4YLTekW3xWIslHXNOi/IsoLFYsHR8SGPDg45W2YUjcZKH6TC0NkZYLBaIfwAIQVSBuRFhac85/3atkO7btuWpqlptWY8GVNWlVMTCxfE5gaW4UAGr9crqrLk6pUrTCYTqqqiaQqEcDRulufUTcPp2RlBEOAHAdaY7tooaOqGvat7nJ60wySQEAJPSQLf+T9fVva4wQYbPAVP8db8/L/mcbLzsyx/8XP9+xeJzKcRo88bkPWLeEx+ket+vg347NvyNPyi5+NXj947/GLbgiiOefOtt5xXaVFw585dwiikqkpOTk6w1rJer3n0yJGjvu8PoVGnp6e0rXumHx4eDJP7Lqg0Ho5Vmk446SyzrDUslwvy3Hnax3HMcrmgaRqWywXGmIGQdc9Q92zc3t4mipxnaxS5YNUbN27wjW98o7PJ6s/Nr14pvsEGG2ywwQYbfHkQ3YyxFRKDQBpBJN0YV9uWK5Md/s0f/BG//Tu/w6P1GQeHpxzef0i9WlGXNfNHR7RGowKfIIxRwqPVFVoKDLCcZ47TEgIjFdrAepGRLzPyvET5UDZLdq/tMLsx5be/+TvsT64wGSeoyNJQoaSPr8GzFqMbhCcwXQfVIpCIT1hSwVdbB/BcZKqQgLBUdUlP8gkhu5n3g4HQs11wje1K48fpiOl0Suj7SAlVWVA3NXEUk2UZnvKoq5Yoipy/6WRCkqaUp6fs7u4SRRGnJ0e0bUOZF4RR9NiAqFer9knzvbIOHCkZhhG+50hXR6g6xWr/OdX5ktZ1jTYGOXi2GqS0yO7UKeEaUdM0lHVD03R+qMaRuevlijSOqeqaMIyYjMeEQej8LztCFuGChepWM18s2F5tsbuzDUKQ5/lA9OhhP5xCMU6c2rOtG6bT6aDoDTvCqSgKqqqmbS2+r7CWwec1jCKuX7/O4dEJ9+4/wFjbqfEaTk5PCUOfR48eEfqeC/yxhjzPXDJ629JoQ+B7+IFP0AQYa4llCFiasxatO+8u40hDaw2e5xPF7vy6gKcvD3mZ4a0dceh8xVyQRJqm0CXUK+URJQltXdO2iijyaZuWoiiwli7IqXUkn3TketNodGswFpTn2myrDdZ0amKpoCNle3JOCIHvOeJf9l4f9OomSRj6XXme7tq0wPe8wSeV7jxGkVOgmLZ1Xr5CuuuREGEVbVMzzysW65y7Dw9IooDxKGaSJIzHKWkSEYY+gR/gewFKSZQSboLHCnSrnX9xd03UTcsqK1isVswXS86WS5Z5QdPZDgihENJZYUgp8ZWP08AIrNFoa5FCUZQVy+WaceKTrZbs7e2hq4ogCFiu17S6RZlgsOLIsmJo043WhGHYKdk1TVUym01p65o4CjubEY2xkuXS+fr23sdKKdZFjpTC+bBuj/C6iRff99nb2+Pg4AA/CPCDiFZrVl3A3gYbbPCbhb4/8jSl4rPe+7T1/rJ/fxoh+ZlJMPt4iJT7+cWpX93XPNuf9teHyBMIYR+bVEvSlLe+/hZN03T9AteHe/ToEdPpdAj17D+TZdlAfi6Xrm+mdcPx8QlbW9tDJVV/vLR2xOh0OnXPdtPg+SmPDh6wNdshDJ19gPMLr0nTEa+++gpBEHRq1og+PHXYiyfUt590xz3f3w022GCDDTbY4DcTfUaKsoBydpwCx19IJFiYhAljP+T1nReoX/kmxmiKquDk7ISj4yOOj4/4+M7HCOHEjt/4we/x7d/+Hn/2n/8z2XyJsjAZT3jh9gv813fepagrXn3jTV579WWu7++wszMhDkNCFeAZhbJgjcHDR0sFUtAawFPONrHrrggLRggQFtFrM38N8FxkahAEBEHQEZG996Pl6OiI+XxOVVe4dAOnbDPG4Pse0+mUJI4w1qCEoq3rC0E8iiiKEUJw/fp10jR1s/hAURbD9/WeqlXbkPij8xJqYxyr3Znx916o/d9FR25F3bY3XUCT7jwqXXjWeem17f5r29a9Z50FQFM3iMGxwQ6qwnOFq6Rpmq683HljjccjRiNXqmW771JWYoXrlGdZxnq1ZjoZoyI1lCE3TTNsJ8IRq+ssAwFlV64VBAGe59G2LWEYdn6rlqa1HXmrUUoNHl9JEvHCC7dZrtacnM2HfS+rivl8ThpHpEnM9f2rpGnKYj4fPG+rqgQhCXwPE0Uo5chnf+ajDSzXGW1d0WqNFIKmaVitV/iBT5Ikg4frl4XRaMRyuWA8GjGbTVguV+5cSUUYRtA0SOWhlEepCzcoke4K9gOfKHZWE1Ip8iynaRvAIqQkCH0W8znCWhCqC4byMFWDUHI4rrIjGZumQfkB3TwEphuQ9QnHrp26EDQXkOG0q1IIdNtirFN2W2ugI6uDIEALHLGpJNoorAqQocK2DVVbU2U589UKzzo1eegrPCWJgshdx56H73sgnPWE6YLg2ralbhuqRlNUDUVd02iDBlqhAIGS4ML1DALrrgkrOzWqdNdfq5FSUxQNh0dHvPbyLXwP8jyn1Zrj09PBDqGfPBBSEscRbduS5znjyYRQBlRViTVO6Tufz53/rDEYq+nV/0HoM52OB/W5U8AagjBiMhnTNI6QTpKEyWTiVLFti+/5tNp0HqwbZeoGG/ym4kli74skGX+VuJAztcFngLWPk+59lZDvB3heMPDT29s7AMNzKkkSHj58gLXu2ZznOTs7O4Ov/WjkKnPKsmR7excArVvG4xFnZ2ckScx4Mur+vs3+/j5hELGzs8N6vXY2T52C+jLPU3emH3//fOLgWSlgvyajkw022GCDDTbY4DND4JSftrVgDUjhhFpKoC1uDC8EiIowFFgjSaIJk8mE27dfpNYN62xF01bopmU7nRFPRgR/+AeUZYkHTEZjojjmpZeuYYTg2tVtbGuIkxmgMEZjadBejbEgpQ82IDBdUX9nvymN6bSoAi2hVgLZ+b7aXxNCVfymDiw22GCDDTbYYIMNNthggw022GCDDTbYYIMNPk9cNt29wQYbbLDBBhtssMEGG2ywwQYbbLDBBhtssMET2JCpG2ywwQYbbLDBBhtssMEGG2ywwQYbbLDBBs+BDZm6wQYbbLDBBhtssMEGG2ywwQYbbLDBBhts8BzYkKkbbLDBBhtssMEGG2ywwQYbbLDBBhtssMEGz4ENmbrBBhtssMEGG2ywwQYbbLDBBhtssMEGG2zwHNiQqRtssMEGG2ywwQYbbLDBBhtssMEGG2ywwQbPgf8/KWEAtl0wskcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1728x216 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "interp = ClassificationInterpretation.from_learner(best_learn)\n",
    "interp.plot_confusion_matrix()\n",
    "interp.plot_top_losses(8, nrows=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "f046c37c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-10-11T18:19:07.762929Z",
     "iopub.status.busy": "2022-10-11T18:19:07.762464Z",
     "iopub.status.idle": "2022-10-11T18:19:09.655814Z",
     "shell.execute_reply": "2022-10-11T18:19:09.654732Z",
     "shell.execute_reply.started": "2022-10-11T18:19:07.762883Z"
    }
   },
   "outputs": [],
   "source": [
    "best_learn.export(\"best_model.pkl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c6eb80b0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}