LeNet-5_MNIST_LTH.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Lottery Ticket Hypothesis: _LeNet-5_ Convolutional Neural Network for MNIST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "import math\n",
    "import tensorflow_model_optimization as tfmot\n",
    "from tensorflow_model_optimization.sparsity import keras as sparsity\n",
    "# from tensorflow.keras import datasets, layers, models\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.keras.layers import AveragePooling2D, Conv2D, MaxPooling2D, ReLU\n",
    "from tensorflow.keras import models, layers, datasets\n",
    "from tensorflow.keras.layers import Dense, Flatten, Reshape, Input, InputLayer\n",
    "from tensorflow.keras.models import Sequential, Model\n",
    "from tensorflow.keras.initializers import RandomNormal\n",
    "# import math\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.0.0'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "env: CUDA_DEVICE_ORDER=PCI_BUS_ID\n",
      "env: CUDA_VISIBLE_DEVICES=2\n"
     ]
    }
   ],
   "source": [
    "%env CUDA_DEVICE_ORDER=PCI_BUS_ID\n",
    "%env CUDA_VISIBLE_DEVICES=2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "batch_size = 60\n",
    "num_classes = 10\n",
    "num_epochs = 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data preprocessing and cleadning:\n",
    "# input image dimensions\n",
    "img_rows, img_cols = 28, 28\n",
    "\n",
    "# Load MNIST dataset-\n",
    "(X_train, y_train), (X_test, y_test) = tf.keras.datasets.mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "'input_shape' which will be used = (28, 28, 1)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "if tf.keras.backend.image_data_format() == 'channels_first':\n",
    "    X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)\n",
    "    X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)\n",
    "    input_shape = (1, img_rows, img_cols)\n",
    "else:\n",
    "    X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)\n",
    "    X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)\n",
    "    input_shape = (img_rows, img_cols, 1)\n",
    "\n",
    "print(\"\\n'input_shape' which will be used = {0}\\n\".format(input_shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert datasets to floating point types-\n",
    "X_train = X_train.astype('float32')\n",
    "X_test = X_test.astype('float32')\n",
    "\n",
    "# Normalize the training and testing datasets-\n",
    "X_train /= 255.0\n",
    "X_test /= 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert class vectors/target to binary class matrices or one-hot encoded values-\n",
    "y_train = tf.keras.utils.to_categorical(y_train, num_classes)\n",
    "y_test = tf.keras.utils.to_categorical(y_test, num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Dimensions of training and testing sets are:\n",
      "X_train.shape = (60000, 28, 28, 1), y_train = (60000, 10)\n",
      "X_test.shape = (10000, 28, 28, 1), y_test = (10000, 10)\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nDimensions of training and testing sets are:\")\n",
    "print(\"X_train.shape = {0}, y_train = {1}\".format(X_train.shape, y_train.shape))\n",
    "print(\"X_test.shape = {0}, y_test = {1}\".format(X_test.shape, y_test.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Prepare MNIST dataset for _GradientTape_ training:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create training and testing datasets-\n",
    "train_dataset = tf.data.Dataset.from_tensor_slices((X_train, y_train))\n",
    "test_dataset = tf.data.Dataset.from_tensor_slices((X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataset = train_dataset.shuffle(buffer_size = 20000, reshuffle_each_iteration = True).batch(batch_size = batch_size, drop_remainder = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dataset = test_dataset.batch(batch_size=batch_size, drop_remainder=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Choose an optimizer and loss function for training-\n",
    "loss_fn = tf.keras.losses.CategoricalCrossentropy()\n",
    "optimizer = tf.keras.optimizers.Adam(lr = 0.0012)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select metrics to measure the error & accuracy of model.\n",
    "# These metrics accumulate the values over epochs and then\n",
    "# print the overall result-\n",
    "train_loss = tf.keras.metrics.Mean(name = 'train_loss')\n",
    "train_accuracy = tf.keras.metrics.CategoricalAccuracy(name = 'train_accuracy')\n",
    "\n",
    "test_loss = tf.keras.metrics.Mean(name = 'test_loss')\n",
    "test_accuracy = tf.keras.metrics.CategoricalAccuracy(name = 'test_accuracy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'end_step parameter' for this dataset =  100000\n"
     ]
    }
   ],
   "source": [
    "# The model is first trained without any pruning for 'num_epochs' epochs-\n",
    "epochs = num_epochs\n",
    "\n",
    "num_train_samples = X_train.shape[0]\n",
    "\n",
    "end_step = np.ceil(1.0 * num_train_samples / batch_size).astype(np.int32) * epochs\n",
    "\n",
    "print(\"'end_step parameter' for this dataset =  {0}\".format(end_step))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify the parameters to be used for layer-wise pruning, NO PRUNING is done here:\n",
    "pruning_params_unpruned = {\n",
    "    'pruning_schedule': sparsity.ConstantSparsity(\n",
    "        target_sparsity=0.0, begin_step=0,\n",
    "        end_step = end_step, frequency=100\n",
    "    )\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "l = tf.keras.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pruned_nn(pruning_params_conv, pruning_params_fc):\n",
    "    \"\"\"\n",
    "    Function to define the architecture of a neural network model\n",
    "    following LeNet-5 architecture for MNIST dataset and using\n",
    "    provided parameter which are used to prune the model.\n",
    "    \n",
    "    \n",
    "    Input: 'pruning_params' Python 3 dictionary containing parameters which are used for pruning\n",
    "    Output: Returns designed and compiled neural network model\n",
    "    \"\"\"\n",
    "    \n",
    "    pruned_model = Sequential()\n",
    "    \n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        Conv2D(\n",
    "            filters = 6, kernel_size = (3, 3),\n",
    "            activation='relu', kernel_initializer = tf.initializers.GlorotUniform(),\n",
    "            strides = (1, 1), padding = 'valid',\n",
    "            input_shape=(28, 28, 1)\n",
    "        ),\n",
    "        **pruning_params_conv)\n",
    "    )\n",
    "       \n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        MaxPooling2D(\n",
    "            pool_size = (2, 2),\n",
    "            strides = (2, 2)\n",
    "        ),\n",
    "        **pruning_params_conv)\n",
    "    )\n",
    "\n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        Conv2D(\n",
    "            filters = 16, kernel_size = (3, 3),\n",
    "            activation='relu', kernel_initializer = tf.initializers.GlorotUniform(),\n",
    "            strides = (1, 1), padding = 'valid'\n",
    "        ),\n",
    "        **pruning_params_conv)\n",
    "    )\n",
    "    \n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        MaxPooling2D(\n",
    "            pool_size = (2, 2),\n",
    "            strides = (2, 2)\n",
    "        ),\n",
    "        **pruning_params_conv)\n",
    "    )\n",
    "   \n",
    "    pruned_model.add(Flatten())\n",
    "    \n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        Dense(\n",
    "            units = 120, activation='relu',\n",
    "            kernel_initializer = tf.initializers.GlorotUniform()\n",
    "        ),\n",
    "        **pruning_params_fc)\n",
    "    )\n",
    "    \n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        Dense(\n",
    "            units = 84, activation='relu',\n",
    "            kernel_initializer = tf.initializers.GlorotUniform()\n",
    "        ),\n",
    "        **pruning_params_fc)\n",
    "    )\n",
    "    \n",
    "    pruned_model.add(sparsity.prune_low_magnitude(\n",
    "        Dense(\n",
    "            units = 10, activation='softmax'\n",
    "        ),\n",
    "        **pruning_params_fc)\n",
    "    )\n",
    "    \n",
    "\n",
    "    # Compile pruned CNN-\n",
    "    pruned_model.compile(\n",
    "        loss=tf.keras.losses.categorical_crossentropy,\n",
    "        # optimizer='adam',\n",
    "        optimizer=tf.keras.optimizers.Adam(lr = 0.0012),\n",
    "        metrics=['accuracy']\n",
    "    )\n",
    "    \n",
    "    \n",
    "    return pruned_model\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add a pruning step callback to peg the pruning step to the optimizer's\n",
    "# step. Also add a callback to add pruning summaries to tensorboard\n",
    "callback = [\n",
    "             sparsity.UpdatePruningStep(),\n",
    "             # sparsity.PruningSummaries(log_dir = logdir, profile_batch=0),\n",
    "             tf.keras.callbacks.EarlyStopping(\n",
    "                 monitor='val_loss', patience = 3,\n",
    "                 min_delta=0.001\n",
    "             )\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /opt/conda/lib/python3.7/site-packages/tensorflow_model_optimization/python/core/sparsity/keras/pruning_wrapper.py:183: Layer.add_variable (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `layer.add_weight` method instead.\n"
     ]
    }
   ],
   "source": [
    "# Initialize a CNN model-\n",
    "orig_model = pruned_nn(pruning_params_unpruned, pruning_params_unpruned)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load Randomly Initialized weights from before-\n",
    "# orig_model.load_weights(\"LeNet_5_MNIST_Random_Weights.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Strip model of it's pruning parameters\n",
    "orig_model_stripped = sparsity.strip_pruning(orig_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save random weights-\n",
    "orig_model.save_weights(\"LeNet_5_MNIST_Random_Weights.h5\", overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save random weights-\n",
    "orig_model.save_weights(\"LeNet_5_MNIST_Winning_Ticket.h5\", overwrite=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d (Conv2D)              (None, 26, 26, 6)         60        \n",
      "_________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D) (None, 13, 13, 6)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 11, 11, 16)        880       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 5, 5, 16)          0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 400)               0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 120)               48120     \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 84)                10164     \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 10)                850       \n",
      "=================================================================\n",
      "Total params: 60,074\n",
      "Trainable params: 60,074\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Get CNN summary-\n",
    "orig_model_stripped.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Sanity check: Whether defined CNN works as intended-\n",
    "'''\n",
    "history = orig_model.fit(\n",
    "    x = X_train, y = y_train,\n",
    "    epochs=num_epochs, batch_size=batch_size,\n",
    "    verbose=1, shuffle=True,\n",
    "    callbacks=callback,\n",
    "    validation_data=(X_test, y_test)\n",
    ")\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.06896029367350129, 0.9828]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# orig_model.evaluate(X_test, y_test, verbose = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Layer-wise pruning:\n",
    "The Lottery Ticket Hypothesis follows a _layer-wise pruning_ heuristic, where the fully-connected (fc) layers are pruned at the rate of 20% per iterative\n",
    "pruning round and the convolutional (conv) layers are pruned at the rate of 10% per iterative pruning round.\n",
    "\n",
    "Therefore, there is a need to compute the percentage of weights being pruned for each layer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Total number of trainable parameters = 60074\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# number of convolutional parameters-\n",
    "conv1 = 60\n",
    "conv2 = 880\n",
    "\n",
    "# number of fully-connected dense parameters-\n",
    "dense1 = 48120\n",
    "dense2 = 10164\n",
    "op_layer = 850\n",
    "\n",
    "\n",
    "# total number of parameters-\n",
    "total_params = conv1 + conv2 + dense1 + dense2 + op_layer\n",
    "\n",
    "print(\"\\nTotal number of trainable parameters = {0}\\n\".format(total_params))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# maximum pruning performed is till 0.5% of all parameters-\n",
    "max_pruned_params = 0.005 * total_params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "loc_tot_params = total_params\n",
    "loc_conv1 = conv1\n",
    "loc_conv2 = conv2\n",
    "loc_dense1 = dense1\n",
    "loc_dense2 = dense2\n",
    "loc_op_layer = op_layer\n",
    "\n",
    "# variable to count number of pruning rounds-\n",
    "n = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lists to hold percentage of weights pruned in each round for all layers in CNN-\n",
    "conv1_pruning = []\n",
    "conv2_pruning = []\n",
    "dense1_pruning = []\n",
    "dense2_pruning = []\n",
    "op_layer_pruning = []\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Conv1 = 54.000, Conv2 = 792.000\n",
      "Dense1 = 38496.000, Dense2 = 8131.200 & O/p layer = 680.000\n",
      "Total number of parameters = 48153.200\n",
      "\n",
      "\n",
      "Conv1 = 48.600, Conv2 = 712.800\n",
      "Dense1 = 30796.800, Dense2 = 6504.960 & O/p layer = 544.000\n",
      "Total number of parameters = 38607.160\n",
      "\n",
      "\n",
      "Conv1 = 43.740, Conv2 = 641.520\n",
      "Dense1 = 24637.440, Dense2 = 5203.968 & O/p layer = 435.200\n",
      "Total number of parameters = 30961.868\n",
      "\n",
      "\n",
      "Conv1 = 39.366, Conv2 = 577.368\n",
      "Dense1 = 19709.952, Dense2 = 4163.174 & O/p layer = 348.160\n",
      "Total number of parameters = 24838.020\n",
      "\n",
      "\n",
      "Conv1 = 35.429, Conv2 = 519.631\n",
      "Dense1 = 15767.962, Dense2 = 3330.540 & O/p layer = 278.528\n",
      "Total number of parameters = 19932.090\n",
      "\n",
      "\n",
      "Conv1 = 31.886, Conv2 = 467.668\n",
      "Dense1 = 12614.369, Dense2 = 2664.432 & O/p layer = 222.822\n",
      "Total number of parameters = 16001.178\n",
      "\n",
      "\n",
      "Conv1 = 28.698, Conv2 = 420.901\n",
      "Dense1 = 10091.495, Dense2 = 2131.545 & O/p layer = 178.258\n",
      "Total number of parameters = 12850.898\n",
      "\n",
      "\n",
      "Conv1 = 25.828, Conv2 = 378.811\n",
      "Dense1 = 8073.196, Dense2 = 1705.236 & O/p layer = 142.606\n",
      "Total number of parameters = 10325.678\n",
      "\n",
      "\n",
      "Conv1 = 23.245, Conv2 = 340.930\n",
      "Dense1 = 6458.557, Dense2 = 1364.189 & O/p layer = 114.085\n",
      "Total number of parameters = 8301.006\n",
      "\n",
      "\n",
      "Conv1 = 20.921, Conv2 = 306.837\n",
      "Dense1 = 5166.846, Dense2 = 1091.351 & O/p layer = 91.268\n",
      "Total number of parameters = 6677.223\n",
      "\n",
      "\n",
      "Conv1 = 18.829, Conv2 = 276.153\n",
      "Dense1 = 4133.477, Dense2 = 873.081 & O/p layer = 73.014\n",
      "Total number of parameters = 5374.554\n",
      "\n",
      "\n",
      "Conv1 = 16.946, Conv2 = 248.538\n",
      "Dense1 = 3306.781, Dense2 = 698.465 & O/p layer = 58.412\n",
      "Total number of parameters = 4329.141\n",
      "\n",
      "\n",
      "Conv1 = 15.251, Conv2 = 223.684\n",
      "Dense1 = 2645.425, Dense2 = 558.772 & O/p layer = 46.729\n",
      "Total number of parameters = 3489.861\n",
      "\n",
      "\n",
      "Conv1 = 13.726, Conv2 = 201.316\n",
      "Dense1 = 2116.340, Dense2 = 447.017 & O/p layer = 37.383\n",
      "Total number of parameters = 2815.783\n",
      "\n",
      "\n",
      "Conv1 = 12.353, Conv2 = 181.184\n",
      "Dense1 = 1693.072, Dense2 = 357.614 & O/p layer = 29.907\n",
      "Total number of parameters = 2274.130\n",
      "\n",
      "\n",
      "Conv1 = 11.118, Conv2 = 163.066\n",
      "Dense1 = 1354.458, Dense2 = 286.091 & O/p layer = 23.925\n",
      "Total number of parameters = 1838.658\n",
      "\n",
      "\n",
      "Conv1 = 10.006, Conv2 = 146.759\n",
      "Dense1 = 1083.566, Dense2 = 228.873 & O/p layer = 19.140\n",
      "Total number of parameters = 1488.345\n",
      "\n",
      "\n",
      "Conv1 = 9.006, Conv2 = 132.083\n",
      "Dense1 = 866.853, Dense2 = 183.098 & O/p layer = 15.312\n",
      "Total number of parameters = 1206.352\n",
      "\n",
      "\n",
      "Conv1 = 8.105, Conv2 = 118.875\n",
      "Dense1 = 693.482, Dense2 = 146.479 & O/p layer = 12.250\n",
      "Total number of parameters = 979.191\n",
      "\n",
      "\n",
      "Conv1 = 7.295, Conv2 = 106.987\n",
      "Dense1 = 554.786, Dense2 = 117.183 & O/p layer = 9.800\n",
      "Total number of parameters = 796.051\n",
      "\n",
      "\n",
      "Conv1 = 6.565, Conv2 = 96.289\n",
      "Dense1 = 443.829, Dense2 = 93.746 & O/p layer = 7.840\n",
      "Total number of parameters = 648.269\n",
      "\n",
      "\n",
      "Conv1 = 5.909, Conv2 = 86.660\n",
      "Dense1 = 355.063, Dense2 = 74.997 & O/p layer = 6.272\n",
      "Total number of parameters = 528.900\n",
      "\n",
      "\n",
      "Conv1 = 5.318, Conv2 = 77.994\n",
      "Dense1 = 284.050, Dense2 = 59.998 & O/p layer = 5.018\n",
      "Total number of parameters = 432.377\n",
      "\n",
      "\n",
      "Conv1 = 4.786, Conv2 = 70.194\n",
      "Dense1 = 227.240, Dense2 = 47.998 & O/p layer = 4.014\n",
      "Total number of parameters = 354.233\n",
      "\n",
      "\n",
      "Conv1 = 4.307, Conv2 = 63.175\n",
      "Dense1 = 181.792, Dense2 = 38.399 & O/p layer = 3.211\n",
      "Total number of parameters = 290.884\n",
      "\n"
     ]
    }
   ],
   "source": [
    "while loc_tot_params >= max_pruned_params:\n",
    "    loc_conv1 *= 0.9    # 10% weights are pruned\n",
    "    loc_conv2 *= 0.9    # 10% weights are pruned\n",
    "    loc_dense1 *= 0.8   # 20% weights are pruned\n",
    "    loc_dense2 *= 0.8   # 20% weights are pruned\n",
    "    loc_op_layer *= 0.8 # 20% weights are pruned\n",
    "    \n",
    "    conv1_pruning.append(((conv1 - loc_conv1) / conv1) * 100)\n",
    "    conv2_pruning.append(((conv2 - loc_conv2) / conv2) * 100)\n",
    "    dense1_pruning.append(((dense1 - loc_dense1) / dense1) * 100)\n",
    "    dense2_pruning.append(((dense2 - loc_dense2) / dense2) * 100)\n",
    "    op_layer_pruning.append(((op_layer - loc_op_layer) / op_layer) * 100)\n",
    "\n",
    "    loc_tot_params = loc_conv1 + loc_conv2 + loc_dense1 + loc_dense2 + loc_op_layer\n",
    "\n",
    "    n += 1\n",
    "\n",
    "    \n",
    "    print(\"\\nConv1 = {0:.3f}, Conv2 = {1:.3f}\".format(loc_conv1, loc_conv2))\n",
    "    print(\"Dense1 = {0:.3f}, Dense2 = {1:.3f} & O/p layer = {2:.3f}\".format(\n",
    "        loc_dense1, loc_dense2, loc_op_layer))\n",
    "    print(\"Total number of parameters = {0:.3f}\\n\".format(loc_tot_params))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "number of pruning rounds = 25\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nnumber of pruning rounds = {0}\\n\\n\".format(n))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_pruning_rounds = n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert from list to np.array-\n",
    "conv1_pruning = np.array(conv1_pruning)\n",
    "conv2_pruning = np.array(conv2_pruning)\n",
    "dense1_pruning = np.array(dense1_pruning)\n",
    "dense2_pruning = np.array(dense2_pruning)\n",
    "op_layer_pruning = np.array(op_layer_pruning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Round off numpy arrays to 3 decimal digits-\n",
    "conv1_pruning = np.round(conv1_pruning, decimals=3)\n",
    "conv2_pruning = np.round(conv2_pruning, decimals=3)\n",
    "dense1_pruning = np.round(dense1_pruning, decimals=3)\n",
    "dense2_pruning = np.round(dense2_pruning, decimals=3)\n",
    "op_layer_pruning = np.round(op_layer_pruning, decimals=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NOTE:\n",
    "\n",
    "For _tensorflow_model_optimization_ to prune, the sparsity has to be between 0 and 1. Therefore, the pruning sparsity has to be divided by 100."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "conv1_pruning = conv1_pruning / 100\n",
    "conv2_pruning = conv2_pruning / 100\n",
    "dense1_pruning = dense1_pruning / 100\n",
    "dense2_pruning = dense2_pruning / 100\n",
    "op_layer_pruning = op_layer_pruning / 100\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.2    , 0.36   , 0.488  , 0.5904 , 0.67232, 0.73786, 0.79028,\n",
       "       0.83223, 0.86578, 0.89263, 0.9141 , 0.93128, 0.94502, 0.95602,\n",
       "       0.96482, 0.97185, 0.97748, 0.98199, 0.98559, 0.98847, 0.99078,\n",
       "       0.99262, 0.9941 , 0.99528, 0.99622])"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dense1_pruning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a mask:\n",
    "A mask is created where all parameters equal to one.\n",
    "\n",
    "This will be used later by _GradientTape_ for training the defined neural network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiate a new neural network model for which, the mask is to be created,\n",
    "# according to the paper-\n",
    "mask_model = pruned_nn(pruning_params_unpruned, pruning_params_unpruned)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Strip the model of its pruning parameters-\n",
    "mask_model_stripped = sparsity.strip_pruning(mask_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Assign all masks to one-\n",
    "\n",
    "for wts in mask_model_stripped.trainable_weights:\n",
    "    wts.assign(\n",
    "        tf.ones_like(\n",
    "            input = wts,\n",
    "            dtype = tf.float32\n",
    "        )\n",
    "\n",
    "    )\n",
    "    # wts.assign(1.)\n",
    "    # wts.assign(tf.where(tf.equal(wts, 0.), 0., 1.))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Mask model metrics:\n",
      "layer-wise number of nonzero parameters in each layer are: \n",
      "\n",
      "54\n",
      "6\n",
      "864\n",
      "16\n",
      "48000\n",
      "120\n",
      "10080\n",
      "84\n",
      "840\n",
      "10\n",
      "\n",
      "Total number of trainable parameters = 60074\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nMask model metrics:\")\n",
    "print(\"layer-wise number of nonzero parameters in each layer are: \\n\")\n",
    "\n",
    "masked_sum_params = 0\n",
    "\n",
    "for layer in mask_model_stripped.trainable_weights:\n",
    "    print(tf.math.count_nonzero(layer, axis = None).numpy())\n",
    "    masked_sum_params += tf.math.count_nonzero(layer, axis = None).numpy()\n",
    "\n",
    "print(\"\\nTotal number of trainable parameters = {0}\\n\".format(masked_sum_params))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "number of pruning rounds for LeNet-5 = 25 and number of epochs = 100\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"\\nnumber of pruning rounds for LeNet-5 = {0} and number of epochs = {1}\\n\".format(num_pruning_rounds, num_epochs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Python 3 dictionary to hold model training metrics for each of _n_ rounds:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Method - 1: Nested Python 3 dictionaries:\n",
    "history_main = {}\n",
    "\n",
    "# for x in range(num_pruning_rounds + 1):\n",
    "for x in range(num_pruning_rounds):\n",
    "    history = {}\n",
    "    \n",
    "    # CNN model, scalar metrics-\n",
    "    history['accuracy'] = np.zeros(shape = num_epochs)\n",
    "    history['val_accuracy'] = np.zeros(shape = num_epochs)\n",
    "    history['loss'] = np.zeros(shape = num_epochs)\n",
    "    history['val_loss'] = np.zeros(shape = num_epochs)\n",
    "    \n",
    "    # compute % of weights pruned at the end of each iterative pruning round-\n",
    "    history['percentage_wts_pruned'] = 10\n",
    "    \n",
    "    # history['epoch_length'] = np.zeros(shape = num_epochs)\n",
    "\n",
    "    history_main[x + 1] = history\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sanity check-\n",
    "history_main.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100,)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sanity check-\n",
    "history_main[10]['accuracy'].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "orig_sum_params = 60074"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# User input parameters for Early Stopping in manual implementation-\n",
    "minimum_delta = 0.001\n",
    "patience = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "best_val_loss = 100\n",
    "loc_patience = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\n",
      "Iterative pruning round: 1\n",
      "\n",
      "\n",
      "Epoch 1, Loss: 0.2187, Accuracy: 93.3917, Test Loss: 0.0651, Test Accuracy: 97.820000\n",
      "Total number of trainable parameters = 60072\n",
      "\n",
      "Epoch 2, Loss: 0.0723, Accuracy: 97.7417, Test Loss: 0.0618, Test Accuracy: 97.869995\n",
      "Total number of trainable parameters = 60074\n",
      "\n",
      "Epoch 3, Loss: 0.0527, Accuracy: 98.3350, Test Loss: 0.0472, Test Accuracy: 98.439995\n",
      "Total number of trainable parameters = 60074\n",
      "\n",
      "Epoch 4, Loss: 0.0427, Accuracy: 98.6233, Test Loss: 0.0356, Test Accuracy: 98.849998\n",
      "Total number of trainable parameters = 60074\n",
      "\n",
      "Epoch 5, Loss: 0.0341, Accuracy: 98.9067, Test Loss: 0.0414, Test Accuracy: 98.580002\n",
      "Total number of trainable parameters = 60074\n",
      "\n",
      "Epoch 6, Loss: 0.0286, Accuracy: 99.0533, Test Loss: 0.0357, Test Accuracy: 98.900002\n",
      "Total number of trainable parameters = 60074\n",
      "\n",
      "Epoch 7, Loss: 0.0235, Accuracy: 99.2433, Test Loss: 0.0377, Test Accuracy: 98.790001\n",
      "Total number of trainable parameters = 60074\n",
      "\n",
      "\n",
      "'EarlyStopping' called!\n",
      "\n",
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/100\n",
      "32940/60000 [===============>..............] - ETA: 5s - loss: 0.0191 - accuracy: 0.9942"
     ]
    }
   ],
   "source": [
    "for i in range(1, num_pruning_rounds + 1):\n",
    "    \n",
    "    print(\"\\n\\n\\nIterative pruning round: {0}\\n\\n\".format(i))\n",
    "    \n",
    "    # Define 'train_one_step()' and 'test_step()' functions here-\n",
    "    @tf.function\n",
    "    def train_one_step(model, mask_model, optimizer, x, y):\n",
    "        '''\n",
    "        Function to compute one step of gradient descent optimization\n",
    "        '''\n",
    "        with tf.GradientTape() as tape:\n",
    "            # Make predictions using defined model-\n",
    "            y_pred = model(x)\n",
    "\n",
    "            # Compute loss-\n",
    "            loss = loss_fn(y, y_pred)\n",
    "        \n",
    "        # Compute gradients wrt defined loss and weights and biases-\n",
    "        grads = tape.gradient(loss, model.trainable_variables)\n",
    "    \n",
    "        # type(grads)\n",
    "        # list\n",
    "    \n",
    "        # List to hold element-wise multiplication between-\n",
    "        # computed gradient and masks-\n",
    "        grad_mask_mul = []\n",
    "    \n",
    "        # Perform element-wise multiplication between computed gradients and masks-\n",
    "        for grad_layer, mask in zip(grads, mask_model.trainable_weights):\n",
    "            grad_mask_mul.append(tf.math.multiply(grad_layer, mask))\n",
    "    \n",
    "        # Apply computed gradients to model's weights and biases-\n",
    "        optimizer.apply_gradients(zip(grad_mask_mul, model.trainable_variables))\n",
    "\n",
    "        # Compute accuracy-\n",
    "        train_loss(loss)\n",
    "        train_accuracy(y, y_pred)\n",
    "\n",
    "        return None\n",
    "    \n",
    "    \n",
    "    @tf.function\n",
    "    def test_step(model, optimizer, data, labels):\n",
    "        \"\"\"\n",
    "        Function to test model performance\n",
    "        on testing dataset\n",
    "        \"\"\"\n",
    "    \n",
    "        predictions = model(data)\n",
    "        t_loss = loss_fn(labels, predictions)\n",
    "\n",
    "        test_loss(t_loss)\n",
    "        test_accuracy(labels, predictions)\n",
    "\n",
    "        return None\n",
    "\n",
    "\n",
    "    # Instantiate a model\n",
    "    model_gt = pruned_nn(pruning_params_unpruned, pruning_params_unpruned)\n",
    "    \n",
    "    # Load winning ticket (from above)-\n",
    "    model_gt.load_weights(\"LeNet_5_MNIST_Winning_Ticket.h5\")\n",
    "    \n",
    "    # Strip model of pruning parameters-\n",
    "    model_gt_stripped = sparsity.strip_pruning(model_gt)\n",
    "    \n",
    "    \n",
    "    # Train model using 'GradientTape'-\n",
    "    \n",
    "    # Initialize parameters for Early Stopping manual implementation-\n",
    "    best_val_loss = 100\n",
    "    loc_patience = 0\n",
    "    \n",
    "    for epoch in range(num_epochs):\n",
    "    \n",
    "        if loc_patience >= patience:\n",
    "            print(\"\\n'EarlyStopping' called!\\n\")\n",
    "            break\n",
    "        \n",
    "        # Reset the metrics at the start of the next epoch\n",
    "        train_loss.reset_states()\n",
    "        train_accuracy.reset_states()\n",
    "        test_loss.reset_states()\n",
    "        test_accuracy.reset_states()\n",
    "        \n",
    "            \n",
    "        for x, y in train_dataset:\n",
    "            # train_one_step(model_gt_stripped, mask_model, optimizer, x, y, grad_mask_mul)\n",
    "            train_one_step(model_gt_stripped, mask_model_stripped, optimizer, x, y)\n",
    "\n",
    "\n",
    "        for x_t, y_t in test_dataset:\n",
    "            # test_step(x_t, y_t)\n",
    "            test_step(model_gt_stripped, optimizer, x_t, y_t)\n",
    "\n",
    "        template = 'Epoch {0}, Loss: {1:.4f}, Accuracy: {2:.4f}, Test Loss: {3:.4f}, Test Accuracy: {4:4f}'\n",
    "    \n",
    "        # 'i' is the index for number of pruning rounds-\n",
    "        history_main[i]['accuracy'][epoch] = train_accuracy.result() * 100\n",
    "        history_main[i]['loss'][epoch] = train_loss.result()\n",
    "        history_main[i]['val_loss'][epoch] = test_loss.result()\n",
    "        history_main[i]['val_accuracy'][epoch] = test_accuracy.result() * 100\n",
    "        \n",
    "        print(template.format(epoch + 1, \n",
    "                              train_loss.result(), train_accuracy.result()*100,\n",
    "                              test_loss.result(), test_accuracy.result()*100))\n",
    "    \n",
    "        # Count number of non-zero parameters in each layer and in total-\n",
    "        # print(\"layer-wise manner model, number of nonzero parameters in each layer are: \\n\")\n",
    "\n",
    "        model_sum_params = 0\n",
    "    \n",
    "        for layer in model_gt_stripped.trainable_weights:\n",
    "            # print(tf.math.count_nonzero(layer, axis = None).numpy())\n",
    "            model_sum_params += tf.math.count_nonzero(layer, axis = None).numpy()\n",
    "    \n",
    "        print(\"Total number of trainable parameters = {0}\\n\".format(model_sum_params))\n",
    "\n",
    "    \n",
    "        # Code for manual Early Stopping:\n",
    "        if np.abs(test_loss.result() < best_val_loss) >= minimum_delta:\n",
    "            # update 'best_val_loss' variable to lowest loss encountered so far-\n",
    "            best_val_loss = test_loss.result()\n",
    "        \n",
    "            # reset 'loc_patience' variable-\n",
    "            loc_patience = 0\n",
    "        \n",
    "        else:  # there is no improvement in monitored metric 'val_loss'\n",
    "            loc_patience += 1  # number of epochs without any improvement\n",
    "\n",
    "    \n",
    "    # Resize numpy arrays according to the epoch when 'EarlyStopping' was called-\n",
    "    for metrics in history_main[i].keys():\n",
    "        history_main[i][metrics] = np.resize(history_main[i][metrics], new_shape = epoch)\n",
    "        # history[metrics] = np.resize(history[metrics], new_shape=epoch)\n",
    "     \n",
    "    \n",
    "    # Save trained model weights-\n",
    "    model_gt.save_weights(\"LeNet_5_MNIST_Trained_Weights.h5\", overwrite=True)\n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "    # Prune trained model:\n",
    "    \n",
    "    # print(\"\\n% of weights to be pruned in round = {0} is: {1:.4f}\\n\".format(i, wts_np[i - 1]))\n",
    "    \n",
    "    # Specify the parameters to be used for layer-wise pruning, Conv layer-\n",
    "    pruning_params_conv = {\n",
    "        'pruning_schedule': sparsity.ConstantSparsity(\n",
    "            target_sparsity=conv1_pruning[i - 1], begin_step = 1000,\n",
    "            end_step = end_step, frequency=100\n",
    "        )\n",
    "    }\n",
    "\n",
    "    \n",
    "    # Specify the parameters to be used for layer-wise pruning, Fully-Connected layer pruning-\n",
    "    pruning_params_fc = {\n",
    "        'pruning_schedule': sparsity.ConstantSparsity(\n",
    "            target_sparsity=dense1_pruning[i - 1], begin_step = 1000,\n",
    "            end_step = end_step, frequency=100\n",
    "        )\n",
    "    }\n",
    "    \n",
    "    \n",
    "    # Instantiate a Nueal Network model to be pruned using parameters from above-\n",
    "    pruned_model = pruned_nn(pruning_params_conv, pruning_params_fc)\n",
    "    \n",
    "    # Load weights from original trained and unpruned model-\n",
    "    pruned_model.load_weights(\"LeNet_5_MNIST_Trained_Weights.h5\")\n",
    "    \n",
    "    # Train pruned NN-\n",
    "    history_pruned = pruned_model.fit(\n",
    "        x = X_train, y = y_train,\n",
    "        batch_size = batch_size,\n",
    "        epochs = epochs,\n",
    "        verbose = 1,\n",
    "        callbacks = callback,\n",
    "        validation_data = (X_test, y_test),\n",
    "        shuffle = True\n",
    "    )\n",
    "    \n",
    "    # Strip the pruning wrappers from pruned model-\n",
    "    pruned_model_stripped = sparsity.strip_pruning(pruned_model)\n",
    "    \n",
    "    # print(\"\\nIn pruned model, number of nonzero parameters in each layer are: \\n\")\n",
    "    pruned_sum_params = 0\n",
    "    \n",
    "    for layer in pruned_model_stripped.trainable_weights:\n",
    "        # print(tf.math.count_nonzero(layer, axis = None).numpy())\n",
    "        pruned_sum_params += tf.math.count_nonzero(layer, axis = None).numpy()\n",
    "    \n",
    "    print(\"\\nRound = {0}, total number of trainable parameters = {1}\\n\".format(i, pruned_sum_params))\n",
    "    # print(\"\\nTotal number of trainable parameters = {0}\\n\".format(pruned_sum_params))\n",
    "    \n",
    "    '''\n",
    "    # Sanity-check: confirm that the weights are actually pruned away from the network-\n",
    "    print(\"\\nRound = {0}, % of weights pruned away = {1:.2f}%\\n\".format( \\\n",
    "                                                i, (orig_sum_params - pruned_sum_params) / orig_sum_params * 100))\n",
    "    '''\n",
    "\n",
    "    # 'i' is the index for number of pruning rounds-\n",
    "    history_main[i]['percentage_wts_pruned'] = ((orig_sum_params - pruned_sum_params) / orig_sum_params) * 100\n",
    " \n",
    "    # Save weights of PRUNED and Trained model BEFORE stripping-\n",
    "    pruned_model.save_weights(\"LeNet_5_MNIST_Pruned_Weights.h5\", overwrite = True)\n",
    "    \n",
    "    \n",
    "    # Create a mask:\n",
    "    \n",
    "    # Instantiate a new neural network model for which, the mask is to be created,\n",
    "    mask_model = pruned_nn(pruning_params_unpruned, pruning_params_unpruned)\n",
    "    \n",
    "    # Load weights of PRUNED model-\n",
    "    mask_model.load_weights(\"LeNet_5_MNIST_Pruned_Weights.h5\")\n",
    "    \n",
    "    # Strip the model of its pruning parameters-\n",
    "    mask_model_stripped = sparsity.strip_pruning(mask_model)\n",
    "    \n",
    "    # For each layer, for each weight which is 0, leave it, as is.\n",
    "    # And for weights which survive the pruning,reinitialize it to ONE (1)-\n",
    "    for wts in mask_model_stripped.trainable_weights:\n",
    "        wts.assign(tf.where(tf.equal(wts, 0.), 0., 1.))\n",
    "    \n",
    "    \n",
    "    \n",
    "    \n",
    "    # Extract Winning Ticket:\n",
    "    \n",
    "    # Instantiate a new neural network model for which, the weights are to be extracted-\n",
    "    winning_ticket_model = pruned_nn(pruning_params_unpruned, pruning_params_unpruned)\n",
    "    \n",
    "    # Load weights of PRUNED model-\n",
    "    winning_ticket_model.load_weights(\"LeNet_5_MNIST_Pruned_Weights.h5\")\n",
    "    \n",
    "    # Strip the model of its pruning parameters-\n",
    "    winning_ticket_model_stripped = sparsity.strip_pruning(winning_ticket_model)\n",
    "    \n",
    "    # For each layer, for each weight which is 0, leave it, as is. And for weights which survive the pruning,\n",
    "    # reinitialize it to the value, the model received BEFORE it was trained and pruned-\n",
    "    for orig_wts, pruned_wts in zip(orig_model_stripped.trainable_weights,\n",
    "                                    winning_ticket_model_stripped.trainable_weights):\n",
    "        pruned_wts.assign(tf.where(tf.equal(pruned_wts, 0), pruned_wts, orig_wts))\n",
    "    \n",
    "    \n",
    "    # Save the weights (with pruning parameters) extracted to a file-\n",
    "    winning_ticket_model.save_weights(\"LeNet_5_MNIST_Winning_Ticket.h5\", overwrite=True)\n",
    "    \n",
    "    \n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"\\nIterative-pruning for LeNet-5 (MNIST) using Lottery Ticket Hypothesis is now complete.\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "os.getcwd()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pickle Python 3 dictionary-\n",
    "with open(\"/home/majumdar/The_Lottery_Ticket_Hypothesis-Resources/LeNet-5_MNIST_history_main_Experiment_3.pkl\", \"wb\") as f:\n",
    "    pickle.dump(history_main, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load pickled Python 3 dictionary-\n",
    "with open(\"/home/majumdar/The_Lottery_Ticket_Hypothesis-Resources/LeNet-5_MNIST_history_main_Experiment_3.pkl\", \"rb\") as f:\n",
    "    history_main = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sanity check-\n",
    "history_main.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualizations for _Iterative Pruning Rounds_:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Python 3 dict for training and testing accuracy visualization-\n",
    "# key is 'percentage_wts_pruned' and value is accuracy\n",
    "plot_accuracy = {}\n",
    "plot_test_accuracy = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# populate 'plot_accuracy'-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['accuracy'])\n",
    "    plot_accuracy[history_main[k]['percentage_wts_pruned']] = history_main[k]['accuracy'][epoch_length - 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "# populate 'plot_test_accuracy'-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['accuracy'])\n",
    "    plot_test_accuracy[history_main[k]['percentage_wts_pruned']] = history_main[k]['val_accuracy'][epoch_length - 1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualization of training and testing accuracy VS percentage of weights\n",
    "# pruned-\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_accuracy.keys()), list(plot_accuracy.values()), label = 'training_accuracy')\n",
    "plt.plot(list(plot_test_accuracy.keys()), list(plot_test_accuracy.values()), label = 'testing_accuracy')\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Accuracy (at convergence) vs. Percentage of weights pruned\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend(loc = 'best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Python 3 dict for training and testing loss visualization-\n",
    "plot_loss = {}\n",
    "plot_test_loss = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "# populate 'plot_loss'-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_loss[history_main[k]['percentage_wts_pruned']] = history_main[k]['loss'][epoch_length - 1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "# populate 'plot_test_loss'-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_test_loss[history_main[k]['percentage_wts_pruned']] = history_main[k]['val_loss'][epoch_length - 1]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualization of training and testing loss VS percentage of remaining weights-\n",
    "# pruned-\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_loss.keys()), list(plot_loss.values()), label = 'training_loss')\n",
    "plt.plot(list(plot_test_loss.keys()), list(plot_test_loss.values()), label = 'testing_loss')\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Loss (at convergence) vs. Percentage of weights pruned\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend(loc = 'best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot number of epochs (Early Stopping) VS percentage of weights pruned-\n",
    "\n",
    "# Python 3 dict to hold number of epochs vs % of weights pruned-\n",
    "plot_num_epochs = {}\n",
    "plot_num_epochs_test = {}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "for k in history_main.keys():\n",
    "\tnum_epochs = len(history_main[k]['accuracy'])\n",
    "\tprint(\"number of epochs = {0}, % of weights remaining = {1:.4f}%\".format(num_epochs, history_main[k]['percentage_wts_pruned']))\n",
    "'''\n",
    "\n",
    "# populate 'plot_num_epochs'-\n",
    "for k in history_main.keys():\n",
    "    num_epochs = len(history_main[k]['accuracy'])\n",
    "    plot_num_epochs[history_main[k]['percentage_wts_pruned']] = num_epochs\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize percentage of weights remaining VS number of epochs (Early Stopping)\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_num_epochs.keys()), list(plot_num_epochs.values()))\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Percentage of weights pruned VS number of epochs (Early Stopping) until convergence\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"number of epochs\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Try to visualize the STARTING training and testing accuracy VS percentage\n",
    "# of remaining weights-\n",
    "\n",
    "# Python 3 dict for visualization-\n",
    "plot_starting_accuracy = {}\n",
    "plot_starting_test_accuracy = {}\n",
    "\n",
    "# Key:    percentage of weights pruned\n",
    "# Value:  starting accuracy\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "# populate 'plot_starting_accuracy'-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_starting_accuracy[history_main[k]['percentage_wts_pruned']] = history_main[k]['accuracy'][0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# populate 'plot_starting_test_accuracy'-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_starting_test_accuracy[history_main[k]['percentage_wts_pruned']] = history_main[k]['val_accuracy'][0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize starting accuracy VS percentage of weights pruned-\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_starting_accuracy.keys()), list(plot_starting_accuracy.values()), label = 'training starting accuracy')\n",
    "plt.plot(list(plot_starting_test_accuracy.keys()), list(plot_starting_test_accuracy.values()), label = 'testing starting accuracy')\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Starting Accuracy vs. Percentage of weights pruned\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend(loc = 'best')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Try to visualize the STARTING training and testing loss VS percentage\n",
    "# of remaining weights-\n",
    "\n",
    "# Python 3 dict for visualization-\n",
    "plot_starting_loss = {}\n",
    "plot_starting_test_loss = {}\n",
    "\n",
    "# Key:\t\tpercentage of weights pruned\n",
    "# Value:\tstarting loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Populate 'plot_starting_loss' Python 3 dict-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_starting_loss[history_main[k]['percentage_wts_pruned']] = history_main[k]['loss'][0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Populate 'plot_starting_test_loss' Python 3 dict-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_starting_test_loss[history_main[k]['percentage_wts_pruned']] = history_main[k]['val_loss'][0]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize Starting training & testing loss VS percentage of weights pruned-\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_starting_loss.keys()), list(plot_starting_loss.values()), label = 'training starting loss')\n",
    "plt.plot(list(plot_starting_test_loss.keys()), list(plot_starting_test_loss.values()), label = 'testing starting loss')\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Starting Loss vs. Percentage of weights pruned\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend(loc = 'best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Try to visualize: percentage of weights remaining VS BEST accuracy within each iterative\n",
    "# pruning round\n",
    "\n",
    "# Python 3 dict for visualization-\n",
    "plot_best_accuracy = {}\n",
    "plot_best_test_accuracy = {}\n",
    "\n",
    "# Key:\t\tpercentage of weights pruned\n",
    "# Value:\tbest accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Populate 'plot_best_accuracy' Python 3 dict-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['accuracy'])\n",
    "    plot_best_accuracy[history_main[k]['percentage_wts_pruned']] = np.amax(history_main[k]['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Populate 'plot_best_test_accuracy' Python 3 dict-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['accuracy'])\n",
    "    plot_best_test_accuracy[history_main[k]['percentage_wts_pruned']] = np.amax(history_main[k]['val_accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize best accuracy VS percentage of weights pruned-\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_best_accuracy.keys()), list(plot_best_accuracy.values()), label = 'training best accuracy')\n",
    "plt.plot(list(plot_best_test_accuracy.keys()), list(plot_best_test_accuracy.values()), label = 'testing best accuracy')\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Best Accuracy vs. Percentage of weights pruned\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend(loc = 'best')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Try to visualize: percentage of weights remaining VS LOWEST loss within each iterative\n",
    "# pruning round\n",
    "\n",
    "# Python 3 dict for visualization-\n",
    "plot_best_loss = {}\n",
    "plot_best_test_loss = {}\n",
    "\n",
    "# Key:\t\tpercentage of weights pruned\n",
    "# Value:\tbest loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Populate 'plot_best_loss' Python 3 dict-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_best_loss[history_main[k]['percentage_wts_pruned']] = np.amin(history_main[k]['loss'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Populate 'plot_best_test_loss' Python 3 dict-\n",
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    plot_best_test_loss[history_main[k]['percentage_wts_pruned']] = np.amin(history_main[k]['val_loss'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize best loss VS percentage of weights pruned-\n",
    "fig=plt.figure(figsize=(10, 9), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "plt.plot(list(plot_best_loss.keys()), list(plot_best_loss.values()), label = 'training best loss')\n",
    "plt.plot(list(plot_best_test_loss.keys()), list(plot_best_test_loss.values()), label = 'testing best loss')\n",
    "\n",
    "plt.title(\"Conv-2 CNN: Best Loss vs. Percentage of weights pruned\")\n",
    "plt.xlabel(\"percentage of weights pruned\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend(loc = 'best')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find the iterative pruning round for which training and testing accuracy is highest:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Iterative round = 4 has highest training accuracy = 99.8000% at percentage of weights pruned = 58.4313%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Find highest training and testing accuracy for all iterative pruning rounds-\n",
    "best_accuracy = 0\n",
    "iterative_round = 0\n",
    "\n",
    "\n",
    "for k in history_main.keys():\n",
    "\tepoch_length = len(history_main[k]['accuracy'])\n",
    "\tif history_main[k]['accuracy'][epoch_length - 1] > best_accuracy:\n",
    "\t\tbest_accuracy = history_main[k]['accuracy'][epoch_length - 1]\n",
    "\t\titerative_round = k\n",
    "\n",
    "\n",
    "print(\"\\nIterative round = {0} has highest training accuracy = {1:.4f}% at percentage of weights pruned = {2:.4f}%\\n\".format(\n",
    "    iterative_round, best_accuracy, history_main[iterative_round]['percentage_wts_pruned']))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Iterative round = 8 has highest testing accuracy = 99.0300% at percentage of weights pruned = 82.4983%\n",
      "\n"
     ]
    }
   ],
   "source": [
    "best_test_accuracy = 0\n",
    "iterative_round = 0\n",
    "\n",
    "\n",
    "for k in history_main.keys():\n",
    "\tepoch_length = len(history_main[k]['val_accuracy'])\n",
    "\tif history_main[k]['val_accuracy'][epoch_length - 1] > best_test_accuracy:\n",
    "\t\tbest_test_accuracy = history_main[k]['val_accuracy'][epoch_length - 1]\n",
    "\t\titerative_round = k\n",
    "\n",
    "\n",
    "print(\"\\nIterative round = {0} has highest testing accuracy = {1:.4f}% at percentage of weights pruned = {2:.4f}%\\n\".format(\n",
    "    iterative_round, best_test_accuracy, history_main[iterative_round]['percentage_wts_pruned']))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "% of wts pruned = 19.7673%, val accuracy = 98.7900%, # of epochs = 7\n",
      "% of wts pruned = 35.5978%, val accuracy = 98.7700%, # of epochs = 8\n",
      "% of wts pruned = 48.2771%, val accuracy = 98.8500%, # of epochs = 8\n",
      "% of wts pruned = 58.4313%, val accuracy = 98.7600%, # of epochs = 8\n",
      "% of wts pruned = 66.5662%, val accuracy = 98.8500%, # of epochs = 6\n",
      "% of wts pruned = 73.0915%, val accuracy = 98.7900%, # of epochs = 5\n",
      "% of wts pruned = 78.3117%, val accuracy = 98.8400%, # of epochs = 6\n",
      "% of wts pruned = 82.4983%, val accuracy = 99.0300%, # of epochs = 7\n",
      "% of wts pruned = 85.8574%, val accuracy = 98.8400%, # of epochs = 5\n",
      "% of wts pruned = 88.5641%, val accuracy = 98.7500%, # of epochs = 6\n",
      "% of wts pruned = 90.7298%, val accuracy = 98.6300%, # of epochs = 7\n",
      "% of wts pruned = 92.4743%, val accuracy = 98.7200%, # of epochs = 9\n",
      "% of wts pruned = 93.8875%, val accuracy = 98.5800%, # of epochs = 9\n",
      "% of wts pruned = 95.0311%, val accuracy = 98.6000%, # of epochs = 7\n",
      "% of wts pruned = 95.9450%, val accuracy = 98.3600%, # of epochs = 10\n",
      "% of wts pruned = 96.6874%, val accuracy = 98.4900%, # of epochs = 13\n",
      "% of wts pruned = 97.2850%, val accuracy = 98.2500%, # of epochs = 17\n",
      "% of wts pruned = 97.7744%, val accuracy = 89.2200%, # of epochs = 12\n",
      "% of wts pruned = 98.1739%, val accuracy = 88.6000%, # of epochs = 24\n",
      "% of wts pruned = 98.4935%, val accuracy = 69.0500%, # of epochs = 18\n",
      "% of wts pruned = 98.7515%, val accuracy = 69.0500%, # of epochs = 24\n",
      "% of wts pruned = 98.9713%, val accuracy = 68.3200%, # of epochs = 37\n",
      "% of wts pruned = 99.1610%, val accuracy = 57.8600%, # of epochs = 28\n",
      "% of wts pruned = 99.2975%, val accuracy = 49.3200%, # of epochs = 25\n",
      "% of wts pruned = 99.4240%, val accuracy = 47.1900%, # of epochs = 19\n"
     ]
    }
   ],
   "source": [
    "for k in history_main.keys():\n",
    "    epoch_length = len(history_main[k]['loss'])\n",
    "    '''\n",
    "    print(\"Iterative pruning round = {0}, val accuracy = {1:.4f}%\".format(\n",
    "        k, history_main[k]['val_accuracy'][epoch_length - 1]))\n",
    "    '''\n",
    "    print(\"% of wts pruned = {0:.4f}%, val accuracy = {1:.4f}%, # of epochs = {2}\".format(\n",
    "        history_main[k]['percentage_wts_pruned'], history_main[k]['val_accuracy'][epoch_length - 1],\n",
    "        len(history_main[k]['loss'])\n",
    "    ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}