RawDataEDA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NJ_220617_WBAN_54779.csv\n",
      "Shape of second file is : (113258, 90)\n",
      "      STATION                    STATION_NAME  ELEVATION  LATITUDE  LONGITUDE  \\\n",
      "0  WBAN:54779  ANDOVER AEROFLEX AIRPORT NJ US      177.7  41.00889  -74.73667   \n",
      "1  WBAN:54779  ANDOVER AEROFLEX AIRPORT NJ US      177.7  41.00889  -74.73667   \n",
      "2  WBAN:54779  ANDOVER AEROFLEX AIRPORT NJ US      177.7  41.00889  -74.73667   \n",
      "3  WBAN:54779  ANDOVER AEROFLEX AIRPORT NJ US      177.7  41.00889  -74.73667   \n",
      "4  WBAN:54779  ANDOVER AEROFLEX AIRPORT NJ US      177.7  41.00889  -74.73667   \n",
      "\n",
      "                 DATE REPORTTPYE  HOURLYSKYCONDITIONS  HOURLYVISIBILITY  \\\n",
      "0 2005-01-01 00:54:00      FM-15                  NaN               NaN   \n",
      "1 2005-01-01 01:54:00      FM-15                  NaN               NaN   \n",
      "2 2005-01-01 02:54:00      FM-15                  NaN               NaN   \n",
      "3 2005-01-01 03:54:00      FM-15                  NaN               NaN   \n",
      "4 2005-01-01 04:54:00      FM-15                  NaN               NaN   \n",
      "\n",
      "  HOURLYPRSENTWEATHERTYPE                ...                  \\\n",
      "0                     NaN                ...                   \n",
      "1                     NaN                ...                   \n",
      "2                     NaN                ...                   \n",
      "3                     NaN                ...                   \n",
      "4                     NaN                ...                   \n",
      "\n",
      "  MonthlyMaxSeaLevelPressureTime MonthlyMinSeaLevelPressureValue  \\\n",
      "0                          -9999                             NaN   \n",
      "1                          -9999                             NaN   \n",
      "2                          -9999                             NaN   \n",
      "3                          -9999                             NaN   \n",
      "4                          -9999                             NaN   \n",
      "\n",
      "  MonthlyMinSeaLevelPressureDate MonthlyMinSeaLevelPressureTime  \\\n",
      "0                          -9999                          -9999   \n",
      "1                          -9999                          -9999   \n",
      "2                          -9999                          -9999   \n",
      "3                          -9999                          -9999   \n",
      "4                          -9999                          -9999   \n",
      "\n",
      "  MonthlyTotalHeatingDegreeDays MonthlyTotalCoolingDegreeDays  \\\n",
      "0                           NaN                           NaN   \n",
      "1                           NaN                           NaN   \n",
      "2                           NaN                           NaN   \n",
      "3                           NaN                           NaN   \n",
      "4                           NaN                           NaN   \n",
      "\n",
      "  MonthlyDeptFromNormalHeatingDD  MonthlyDeptFromNormalCoolingDD  \\\n",
      "0                            NaN                             NaN   \n",
      "1                            NaN                             NaN   \n",
      "2                            NaN                             NaN   \n",
      "3                            NaN                             NaN   \n",
      "4                            NaN                             NaN   \n",
      "\n",
      "  MonthlyTotalSeasonToDateHeatingDD  MonthlyTotalSeasonToDateCoolingDD  \n",
      "0                               NaN                                NaN  \n",
      "1                               NaN                                NaN  \n",
      "2                               NaN                                NaN  \n",
      "3                               NaN                                NaN  \n",
      "4                               NaN                                NaN  \n",
      "\n",
      "[5 rows x 90 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\IPython\\core\\interactiveshell.py:2717: DtypeWarning: Columns (9,10,11,12,13,14,15,16,20,23,24,25,26,27,28,29,33,34,44,45,46,47,71,72,73,84,85,86,87) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n"
     ]
    }
   ],
   "source": [
    "import boto3\n",
    "from boto.s3.connection import S3Connection\n",
    "import os\n",
    "import json\n",
    "import boto.s3\n",
    "import sys\n",
    "import datetime\n",
    "from boto.s3.key import Key\n",
    "from pprint import pprint\n",
    "import pandas as pd\n",
    "import urllib\n",
    "import csv\n",
    "import io\n",
    "import requests\n",
    "import time\n",
    "import json\n",
    "import datetime\n",
    "from pprint import pprint\n",
    "import scipy\n",
    "import numpy as np\n",
    "\n",
    "# Loading the config.json file and get merged Csv\n",
    "\n",
    "with open('config.json') as data_file:    \n",
    "        data = json.load(data_file)   \n",
    "\n",
    "#Extracting Data From last File Created.\n",
    "\n",
    "        lastchangedfile= data[\"lastChangedFile\"]   \n",
    "        print(lastchangedfile)\n",
    "        rawdata= pd.read_csv(lastchangedfile, parse_dates=['DATE'])\n",
    "        print(\"Shape of second file is :\",rawdata.shape)\n",
    "        print(rawdata.head(5))\n",
    "        \n",
    "        \n",
    "  \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Selecting Important Data columns from File "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 DATE HOURLYDRYBULBTEMPF HOURLYDRYBULBTEMPC  \\\n",
      "0 2005-01-01 00:54:00                 48                8.9   \n",
      "1 2005-01-01 01:54:00                 46                7.8   \n",
      "2 2005-01-01 02:54:00                 44                6.7   \n",
      "\n",
      "  HOURLYWETBULBTEMPF HOURLYWETBULBTEMPC HOURLYDewPointTempF  \\\n",
      "0                 44                6.5                  39   \n",
      "1                 43                5.9                  39   \n",
      "2                 41                4.8                  37   \n",
      "\n",
      "  HOURLYDewPointTempC HOURLYRelativeHumidity  HOURLYWindSpeed  \\\n",
      "0                 3.9                     71              7.0   \n",
      "1                 3.9                     77              6.0   \n",
      "2                 2.8                     76              3.0   \n",
      "\n",
      "  HOURLYWindDirection  HOURLYWindGustSpeed HOURLYStationPressure  \\\n",
      "0                 200            19.990065                 29.55   \n",
      "1                 190            19.990065                 29.54   \n",
      "2                 200            19.990065                 29.56   \n",
      "\n",
      "   HOURLYPressureTendency HOURLYSeaLevelPressure HOURLYPrecip  \\\n",
      "0                8.000000                  30.19            0   \n",
      "1                4.276953                  30.19            0   \n",
      "2                4.276953                   30.2            0   \n",
      "\n",
      "  HOURLYAltimeterSetting REPORTTPYE  DAILYSunrise  DAILYSunset  \n",
      "0                  30.18      FM-15           724         1641  \n",
      "1                  30.17      FM-15           724         1641  \n",
      "2                  30.19      FM-15           724         1641  \n",
      "DATE                      datetime64[ns]\n",
      "HOURLYDRYBULBTEMPF                object\n",
      "HOURLYDRYBULBTEMPC                object\n",
      "HOURLYWETBULBTEMPF                object\n",
      "HOURLYWETBULBTEMPC                object\n",
      "HOURLYDewPointTempF               object\n",
      "HOURLYDewPointTempC               object\n",
      "HOURLYRelativeHumidity            object\n",
      "HOURLYWindSpeed                  float64\n",
      "HOURLYWindDirection               object\n",
      "HOURLYWindGustSpeed              float64\n",
      "HOURLYStationPressure             object\n",
      "HOURLYPressureTendency           float64\n",
      "HOURLYSeaLevelPressure            object\n",
      "HOURLYPrecip                      object\n",
      "HOURLYAltimeterSetting            object\n",
      "REPORTTPYE                        object\n",
      "DAILYSunrise                       int64\n",
      "DAILYSunset                        int64\n",
      "dtype: object\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>DATE</th>\n",
       "      <th>HOURLYDRYBULBTEMPF</th>\n",
       "      <th>HOURLYDRYBULBTEMPC</th>\n",
       "      <th>HOURLYWETBULBTEMPF</th>\n",
       "      <th>HOURLYWETBULBTEMPC</th>\n",
       "      <th>HOURLYDewPointTempF</th>\n",
       "      <th>HOURLYDewPointTempC</th>\n",
       "      <th>HOURLYRelativeHumidity</th>\n",
       "      <th>HOURLYWindSpeed</th>\n",
       "      <th>HOURLYWindDirection</th>\n",
       "      <th>HOURLYWindGustSpeed</th>\n",
       "      <th>HOURLYStationPressure</th>\n",
       "      <th>HOURLYPressureTendency</th>\n",
       "      <th>HOURLYSeaLevelPressure</th>\n",
       "      <th>HOURLYPrecip</th>\n",
       "      <th>HOURLYAltimeterSetting</th>\n",
       "      <th>REPORTTPYE</th>\n",
       "      <th>DAILYSunrise</th>\n",
       "      <th>DAILYSunset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2005-01-01 00:54:00</td>\n",
       "      <td>48</td>\n",
       "      <td>8.9</td>\n",
       "      <td>44</td>\n",
       "      <td>6.5</td>\n",
       "      <td>39</td>\n",
       "      <td>3.9</td>\n",
       "      <td>71</td>\n",
       "      <td>7.0</td>\n",
       "      <td>200</td>\n",
       "      <td>19.990065</td>\n",
       "      <td>29.55</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>30.19</td>\n",
       "      <td>0</td>\n",
       "      <td>30.18</td>\n",
       "      <td>FM-15</td>\n",
       "      <td>724</td>\n",
       "      <td>1641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2005-01-01 01:54:00</td>\n",
       "      <td>46</td>\n",
       "      <td>7.8</td>\n",
       "      <td>43</td>\n",
       "      <td>5.9</td>\n",
       "      <td>39</td>\n",
       "      <td>3.9</td>\n",
       "      <td>77</td>\n",
       "      <td>6.0</td>\n",
       "      <td>190</td>\n",
       "      <td>19.990065</td>\n",
       "      <td>29.54</td>\n",
       "      <td>4.276953</td>\n",
       "      <td>30.19</td>\n",
       "      <td>0</td>\n",
       "      <td>30.17</td>\n",
       "      <td>FM-15</td>\n",
       "      <td>724</td>\n",
       "      <td>1641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2005-01-01 02:54:00</td>\n",
       "      <td>44</td>\n",
       "      <td>6.7</td>\n",
       "      <td>41</td>\n",
       "      <td>4.8</td>\n",
       "      <td>37</td>\n",
       "      <td>2.8</td>\n",
       "      <td>76</td>\n",
       "      <td>3.0</td>\n",
       "      <td>200</td>\n",
       "      <td>19.990065</td>\n",
       "      <td>29.56</td>\n",
       "      <td>4.276953</td>\n",
       "      <td>30.2</td>\n",
       "      <td>0</td>\n",
       "      <td>30.19</td>\n",
       "      <td>FM-15</td>\n",
       "      <td>724</td>\n",
       "      <td>1641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2005-01-01 03:54:00</td>\n",
       "      <td>43</td>\n",
       "      <td>6.1</td>\n",
       "      <td>41</td>\n",
       "      <td>4.8</td>\n",
       "      <td>38</td>\n",
       "      <td>3.3</td>\n",
       "      <td>82</td>\n",
       "      <td>3.0</td>\n",
       "      <td>200</td>\n",
       "      <td>19.990065</td>\n",
       "      <td>29.55</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>30.19</td>\n",
       "      <td>0</td>\n",
       "      <td>30.18</td>\n",
       "      <td>FM-15</td>\n",
       "      <td>724</td>\n",
       "      <td>1641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2005-01-01 04:54:00</td>\n",
       "      <td>45</td>\n",
       "      <td>7.2</td>\n",
       "      <td>42</td>\n",
       "      <td>5.4</td>\n",
       "      <td>38</td>\n",
       "      <td>3.3</td>\n",
       "      <td>77</td>\n",
       "      <td>6.0</td>\n",
       "      <td>VRB</td>\n",
       "      <td>19.990065</td>\n",
       "      <td>29.53</td>\n",
       "      <td>4.276953</td>\n",
       "      <td>30.18</td>\n",
       "      <td>0</td>\n",
       "      <td>30.16</td>\n",
       "      <td>FM-15</td>\n",
       "      <td>724</td>\n",
       "      <td>1641</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 DATE HOURLYDRYBULBTEMPF HOURLYDRYBULBTEMPC  \\\n",
       "0 2005-01-01 00:54:00                 48                8.9   \n",
       "1 2005-01-01 01:54:00                 46                7.8   \n",
       "2 2005-01-01 02:54:00                 44                6.7   \n",
       "3 2005-01-01 03:54:00                 43                6.1   \n",
       "4 2005-01-01 04:54:00                 45                7.2   \n",
       "\n",
       "  HOURLYWETBULBTEMPF HOURLYWETBULBTEMPC HOURLYDewPointTempF  \\\n",
       "0                 44                6.5                  39   \n",
       "1                 43                5.9                  39   \n",
       "2                 41                4.8                  37   \n",
       "3                 41                4.8                  38   \n",
       "4                 42                5.4                  38   \n",
       "\n",
       "  HOURLYDewPointTempC HOURLYRelativeHumidity  HOURLYWindSpeed  \\\n",
       "0                 3.9                     71              7.0   \n",
       "1                 3.9                     77              6.0   \n",
       "2                 2.8                     76              3.0   \n",
       "3                 3.3                     82              3.0   \n",
       "4                 3.3                     77              6.0   \n",
       "\n",
       "  HOURLYWindDirection  HOURLYWindGustSpeed HOURLYStationPressure  \\\n",
       "0                 200            19.990065                 29.55   \n",
       "1                 190            19.990065                 29.54   \n",
       "2                 200            19.990065                 29.56   \n",
       "3                 200            19.990065                 29.55   \n",
       "4                 VRB            19.990065                 29.53   \n",
       "\n",
       "   HOURLYPressureTendency HOURLYSeaLevelPressure HOURLYPrecip  \\\n",
       "0                8.000000                  30.19            0   \n",
       "1                4.276953                  30.19            0   \n",
       "2                4.276953                   30.2            0   \n",
       "3                8.000000                  30.19            0   \n",
       "4                4.276953                  30.18            0   \n",
       "\n",
       "  HOURLYAltimeterSetting REPORTTPYE  DAILYSunrise  DAILYSunset  \n",
       "0                  30.18      FM-15           724         1641  \n",
       "1                  30.17      FM-15           724         1641  \n",
       "2                  30.19      FM-15           724         1641  \n",
       "3                  30.18      FM-15           724         1641  \n",
       "4                  30.16      FM-15           724         1641  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df= rawdata[['DATE','HOURLYDRYBULBTEMPF', \n",
    "                                  'HOURLYDRYBULBTEMPC', \n",
    "                                  'HOURLYWETBULBTEMPF', \n",
    "                                  'HOURLYWETBULBTEMPC',\n",
    "                                  'HOURLYDewPointTempF', \n",
    "                                  'HOURLYDewPointTempC', \n",
    "                                  'HOURLYRelativeHumidity', \n",
    "                                  'HOURLYWindSpeed', \n",
    "                                  'HOURLYWindDirection', \n",
    "                                  'HOURLYWindGustSpeed', \n",
    "                                  'HOURLYStationPressure',\n",
    "                                  'HOURLYPressureTendency', \n",
    "                                  'HOURLYSeaLevelPressure',\n",
    "                                  'HOURLYPrecip', \n",
    "                                  'HOURLYAltimeterSetting',\n",
    "                                  'REPORTTPYE',\n",
    "                                  'DAILYSunrise',\n",
    "                                  'DAILYSunset']]\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "df = df[df.REPORTTPYE != 'SOD']\n",
    "\n",
    "noNaNdf= df.fillna(scipy.mean(df)) \n",
    "print (noNaNdf.head(3))\n",
    "print(noNaNdf.dtypes)\n",
    "\n",
    "noNaNdf.head(5)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Segregate data into Time, Date, Months, Day and Year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DATE                      datetime64[ns]\n",
      "HOURLYDRYBULBTEMPF                object\n",
      "HOURLYDRYBULBTEMPC                object\n",
      "HOURLYWETBULBTEMPF                object\n",
      "HOURLYWETBULBTEMPC                object\n",
      "HOURLYDewPointTempF               object\n",
      "HOURLYDewPointTempC               object\n",
      "HOURLYRelativeHumidity            object\n",
      "HOURLYWindSpeed                  float64\n",
      "HOURLYWindDirection               object\n",
      "HOURLYWindGustSpeed              float64\n",
      "HOURLYStationPressure             object\n",
      "HOURLYPressureTendency           float64\n",
      "HOURLYSeaLevelPressure            object\n",
      "HOURLYPrecip                      object\n",
      "HOURLYAltimeterSetting            object\n",
      "REPORTTPYE                        object\n",
      "DAILYSunrise                       int64\n",
      "DAILYSunset                        int64\n",
      "new_date                          object\n",
      "new_time                          object\n",
      "year                               int64\n",
      "month                              int64\n",
      "day                                int64\n",
      "dtype: object\n",
      "                 DATE HOURLYDRYBULBTEMPF HOURLYDRYBULBTEMPC  \\\n",
      "0 2005-01-01 00:54:00                 48                8.9   \n",
      "1 2005-01-01 01:54:00                 46                7.8   \n",
      "2 2005-01-01 02:54:00                 44                6.7   \n",
      "\n",
      "  HOURLYWETBULBTEMPF HOURLYWETBULBTEMPC HOURLYDewPointTempF  \\\n",
      "0                 44                6.5                  39   \n",
      "1                 43                5.9                  39   \n",
      "2                 41                4.8                  37   \n",
      "\n",
      "  HOURLYDewPointTempC HOURLYRelativeHumidity  HOURLYWindSpeed  \\\n",
      "0                 3.9                     71              7.0   \n",
      "1                 3.9                     77              6.0   \n",
      "2                 2.8                     76              3.0   \n",
      "\n",
      "  HOURLYWindDirection ...  HOURLYPrecip HOURLYAltimeterSetting  REPORTTPYE  \\\n",
      "0                 200 ...             0                  30.18       FM-15   \n",
      "1                 190 ...             0                  30.17       FM-15   \n",
      "2                 200 ...             0                  30.19       FM-15   \n",
      "\n",
      "  DAILYSunrise DAILYSunset    new_date  new_time  year  month day  \n",
      "0          724        1641  2005-01-01  00:54:00  2005      1   1  \n",
      "1          724        1641  2005-01-01  01:54:00  2005      1   1  \n",
      "2          724        1641  2005-01-01  02:54:00  2005      1   1  \n",
      "\n",
      "[3 rows x 24 columns]\n",
      "DATE                      datetime64[ns]\n",
      "HOURLYDRYBULBTEMPF                object\n",
      "HOURLYDRYBULBTEMPC                object\n",
      "HOURLYWETBULBTEMPF                object\n",
      "HOURLYWETBULBTEMPC                object\n",
      "HOURLYDewPointTempF               object\n",
      "HOURLYDewPointTempC               object\n",
      "HOURLYRelativeHumidity            object\n",
      "HOURLYWindSpeed                  float64\n",
      "HOURLYWindDirection               object\n",
      "HOURLYWindGustSpeed              float64\n",
      "HOURLYStationPressure             object\n",
      "HOURLYPressureTendency           float64\n",
      "HOURLYSeaLevelPressure            object\n",
      "HOURLYPrecip                      object\n",
      "HOURLYAltimeterSetting            object\n",
      "REPORTTPYE                        object\n",
      "DAILYSunrise                       int64\n",
      "DAILYSunset                        int64\n",
      "new_date                          object\n",
      "new_time                          object\n",
      "year                               int64\n",
      "month                              int64\n",
      "day                                int64\n",
      "dtype: object\n"
     ]
    }
   ],
   "source": [
    "df['new_date'] = [d.date() for d in df['DATE']]\n",
    "df['new_time'] = [d.time() for d in df['DATE']]                        \n",
    "df['year'] = df.DATE.dt.year\n",
    "df['month'] = df.DATE.dt.month\n",
    "df['day'] = df.DATE.dt.day\n",
    "print (df.dtypes)\n",
    "df\n",
    "\n",
    "#Removing the NaN values \n",
    "noNaNdf= df.fillna(scipy.mean(df)) \n",
    "print (noNaNdf.head(3))\n",
    "print(noNaNdf.dtypes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trend in Maximum and Minimum Temperatures over the years."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  HOURLYDRYBULBTEMPC  year\n",
      "0                8.9  2005\n",
      "1                7.8  2005\n",
      "2                6.7  2005\n",
      "object\n",
      "int32\n",
      "float64\n",
      "    year  HOURLYDRYBULBTEMPC\n",
      "0   2005                34.4\n",
      "1   2006                35.0\n",
      "2   2007                33.9\n",
      "3   2008                35.0\n",
      "4   2009                32.2\n",
      "5   2010                36.7\n",
      "6   2011                37.8\n",
      "7   2012                36.1\n",
      "8   2013                35.6\n",
      "9   2014                32.2\n",
      "10  2015                33.3\n",
      "11  2016                34.4\n",
      "12  2017                33.9\n",
      "    year  HOURLYDRYBULBTEMPC\n",
      "0   2005               -24.4\n",
      "1   2006               -12.8\n",
      "2   2007               -16.1\n",
      "3   2008               -13.3\n",
      "4   2009               -21.7\n",
      "5   2010               -15.0\n",
      "6   2011               -23.3\n",
      "7   2012               -15.0\n",
      "8   2013               -16.1\n",
      "9   2014               -22.2\n",
      "10  2015               -24.4\n",
      "11  2016               -20.6\n",
      "12  2017               -15.6\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:18: FutureWarning: convert_objects is deprecated.  Use the data-type specific converters pd.to_datetime, pd.to_timedelta and pd.to_numeric.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEKCAYAAAACS67iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9fc93f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAEKCAYAAAARnO4WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqICvKIoSIVTAVxRFiRAq4CuKokSI/w9OGF9E2A96cwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12d2f170>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "#define a dataframe\n",
    "dfYearMaxMinTemp= noNaNdf[[ \"HOURLYDRYBULBTEMPC\", \"year\"]]\n",
    "\n",
    "\n",
    "#Replace NaN by mean Value\n",
    "noNaNdfYearMaxMinTemp= dfYearMaxMinTemp.fillna(scipy.mean(dfYearMaxMinTemp)) \n",
    "print (noNaNdfYearMaxMinTemp.head(3) )\n",
    "\n",
    "#Change the dataframe into integer value\n",
    "noNaNdfYearMaxMinTemp[\"year\"]=noNaNdfYearMaxMinTemp[\"year\"].astype(int)\n",
    "print(noNaNdfYearMaxMinTemp[\"HOURLYDRYBULBTEMPC\"].dtype)\n",
    "\n",
    "\n",
    "#Change the dataframe into numeric(float 64)\n",
    "noNaNdfYearMaxMinTemp[\"HOURLYDRYBULBTEMPC\"]=noNaNdfYearMaxMinTemp[\"HOURLYDRYBULBTEMPC\"].convert_objects(convert_numeric=True)\n",
    "print (noNaNdfYearMaxMinTemp[\"year\"].dtype)\n",
    "\n",
    "#printing the datatype\n",
    "print(noNaNdfYearMaxMinTemp[\"HOURLYDRYBULBTEMPC\"].dtype)\n",
    "\n",
    "#creating Max temperature dataframe over years\n",
    "maxtempdata= noNaNdfYearMaxMinTemp.groupby('year')['HOURLYDRYBULBTEMPC'].max().reset_index()\n",
    "print(maxtempdata)\n",
    "\n",
    "#creating Minimum temperature dataframe over years\n",
    "mintempdata= (noNaNdfYearMaxMinTemp.groupby('year')['HOURLYDRYBULBTEMPC']).min().reset_index()\n",
    "print(mintempdata)\n",
    "\n",
    "#Varialbes for Max\n",
    "maxx=maxtempdata[['year']]\n",
    "maxyy=maxtempdata[['HOURLYDRYBULBTEMPC']]\n",
    "\n",
    "#Variables for Min\n",
    "minx=mintempdata[['year']]\n",
    "minyy=mintempdata[['HOURLYDRYBULBTEMPC']]\n",
    "\n",
    "# Line-plot  Max temperature over years\n",
    "maxtempdata.plot(x=\"year\", y=\"HOURLYDRYBULBTEMPC\", kind='line')\n",
    "\n",
    "\n",
    "#Line-plot min temperatures over years\n",
    "mintempdata.plot(x=\"year\", y=\"HOURLYDRYBULBTEMPC\", kind='line')\n",
    "plt.show()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot for length of day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0   1900-01-01 16:41:00\n",
      "1   1900-01-01 16:41:00\n",
      "2   1900-01-01 16:41:00\n",
      "Name: DAILYSunset, dtype: datetime64[ns]\n",
      "0   1900-01-01 07:24:00\n",
      "1   1900-01-01 07:24:00\n",
      "2   1900-01-01 07:24:00\n",
      "Name: DAILYSunrise, dtype: datetime64[ns]\n",
      "0    9.283333\n",
      "1    9.283333\n",
      "2    9.283333\n",
      "dtype: float64\n",
      "   year  month  daylight\n",
      "0  2005      1  9.283333\n",
      "1  2005      1  9.283333\n",
      "2  2005      1  9.283333\n",
      "3  2005      1  9.283333\n",
      "4  2005      1  9.283333\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:14: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "time_func = lambda x: pd.Timestamp(pd.to_datetime(x, format = '%H%M'))\n",
    "\n",
    "dataforsunrise=noNaNdf['DAILYSunrise'].apply(time_func)\n",
    "dataforsunset=noNaNdf['DAILYSunset'].apply(time_func)\n",
    "daylenght=(dataforsunset-dataforsunrise).astype('timedelta64[m]')/60\n",
    "\n",
    "print(dataforsunset.head(3))\n",
    "print(dataforsunrise.head(3))\n",
    "print(daylenght.head(3))\n",
    "\n",
    "# daylen1= (daylenght.groupby('year', 'month')['hour']).mean().reset_index()\n",
    "# print(daylen1)\n",
    "daylendf=noNaNdf[['year', 'month']]\n",
    "daylendf['daylight']=daylenght.abs()\n",
    "print (daylendf.head(5))\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n",
    "# Yearly max and min bar chart\n",
    "# plt.figure(figsize=(12,8))\n",
    "# sns.barplot(x=\"year\", y=\"temp\", noNaNdfYearMaxMinTemp=monthly.sort_values(['year', 'HOURLYDRYBULBTEMPC']))\n",
    "# plt.xlabel(\"year\")\n",
    "# plt.ylabel(\"Temp\")\n",
    "# plt.title(\"Yearly max and min bar chart {}\".format(station))\n",
    "\n",
    "\n",
    "df1 = pd.DataFrame(index=noNaNdfYearMaxMinTemp.index, columns=list('year'))\n",
    "df1 = df1.cumsum()\n",
    "plt.figure(); \n",
    "df1.plot();\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1232ac10>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "mintempdata = mintempdata.cumsum()\n",
    "mintempdata.plot()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# plt.plot(maxx,maxyy, minx, minyy)\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   year  month HOURLYRelativeHumidity\n",
      "0  2005      1                     71\n",
      "1  2005      1                     77\n",
      "2  2005      1                     76\n",
      "      year  month HOURLYRelativeHumidity\n",
      "4527  2005      7                     87\n",
      "4528  2005      7                     87\n",
      "4529  2005      7                     87\n",
      "4530  2005      7                     87\n",
      "4531  2005      7                     90\n",
      "    year  HOURLYRelativeHumidity\n",
      "0   2005               75.203230\n",
      "1   2006               77.254032\n",
      "2   2007               73.373656\n",
      "3   2008               70.323263\n",
      "4   2009               77.879032\n",
      "5   2010               73.858871\n",
      "6   2011               74.732527\n",
      "7   2012               74.354839\n",
      "8   2013               76.952957\n",
      "9   2014               76.237903\n",
      "10  2015               77.575269\n",
      "11  2016               76.096774\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:5: FutureWarning: convert_objects is deprecated.  Use the data-type specific converters pd.to_datetime, pd.to_timedelta and pd.to_numeric.\n",
      "  \"\"\"\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEZCAYAAACHCd7XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nyQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12304a70>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "janYearlyData=noNaNdf[['year','month', 'HOURLYRelativeHumidity']]\n",
    "print (janYearlyData.head(3))\n",
    "janYearlyData=janYearlyData.loc[janYearlyData['month']==7]\n",
    "print (janYearlyData.head(5))\n",
    "janYearlyData[\"HOURLYRelativeHumidity\"]=janYearlyData[\"HOURLYRelativeHumidity\"].convert_objects(convert_numeric=True)\n",
    "\n",
    "# create dataframe for humidity over years \n",
    "janYearlyData= (janYearlyData.groupby('year')['HOURLYRelativeHumidity']).mean().reset_index()\n",
    "print(janYearlyData)\n",
    "\n",
    "\n",
    "# plot for August Humidity over years \n",
    "janYearlyData.plot(x=\"year\", y=\"HOURLYRelativeHumidity\", kind='bar')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}