github_pull_requests_scores_senticr 2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import os,json\n",
    "import sys\n",
    "import augur\n",
    "import sqlalchemy as s\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "import matplotlib.pyplot as plot\n",
    "import matplotlib.dates as mdates\n",
    "import datetime as dt\n",
    "#!{sys.executable} -m pip install sklearn\n",
    "#!{sys.executable} -m pip install xlrd\n",
    "#!{sys.executable} -m pip install imblearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/keanu/temp/augur_push/augur5/augur/SentiCR\n",
      "/home/keanu/temp/augur_push/augur5/augur/SentiCR/SentiCR\n",
      "Reading data from oracle..\n",
      "Training classifier model..\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/keanu/anaconda3/envs/augur/lib/python3.6/site-packages/sklearn/utils/deprecation.py:77: DeprecationWarning: Function _ratio_float is deprecated; Use a float for 'ratio' is deprecated from version 0.2. The support will be removed in 0.4. Use a dict, str, or a callable instead.\n",
      "  warnings.warn(msg, category=DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['github_issues', 'github_issues_2', 'github_issues_repo_jobs', 'github_issues_sentiment_scores', 'github_issues_sentiment_scores_senticr', 'github_pull_request_repo_jobs', 'github_pull_requests', 'github_pull_requests_2', 'github_pull_requests_sentiment_scores', 'github_pull_requests_sentiment_scores_2', 'issue_response_time', 'issue_response_time2', 'issue_response_time_3', 'mail_lists', 'mail_lists_sentiment_scores', 'mail_lists_sentiment_scores_senticr', 'mailing_list_jobs']\n",
      "   augurmsgID backend_name                      repo_link   owner repo  \\\n",
      "0           1       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "1           2       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "2           3       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "3           4       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "4           5       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "\n",
      "                subject  status      category  pull_request_number  \\\n",
      "0  add meeting schedule  closed  pull_request                   18   \n",
      "1  add meeting schedule  closed  pull_request                   18   \n",
      "2  add meeting schedule  closed  pull_request                   18   \n",
      "3  add meeting schedule  closed  pull_request                   18   \n",
      "4  add meeting schedule  closed  pull_request                   18   \n",
      "\n",
      "                 date  pull_request_id          user  \\\n",
      "0 2016-10-06 04:03:18        181319945         Trott   \n",
      "1 2016-10-06 13:17:53        181319945      mhdawson   \n",
      "2 2016-10-06 21:16:54        181319945  williamkapke   \n",
      "3 2016-10-06 21:31:31        181319945       jasnell   \n",
      "4 2016-10-06 21:34:01        181319945   MylesBorins   \n",
      "\n",
      "                                                body  \n",
      "0  Not sure that the README is the best place for...  \n",
      "1                                             LGTM\\n  \n",
      "2  This is a bit of a tangent but.. should the Me...  \n",
      "3  I would definitely prefer the meeting notices ...  \n",
      "4                       +1 on meeting notices here\\n  \n",
      "      augurmsgID backend_name                          repo_link     owner  \\\n",
      "3778        3779       GitHub  https://github.com/torvalds/linux  torvalds   \n",
      "3779        3780       GitHub  https://github.com/torvalds/linux  torvalds   \n",
      "3780        3781       GitHub  https://github.com/torvalds/linux  torvalds   \n",
      "3781        3782       GitHub  https://github.com/torvalds/linux  torvalds   \n",
      "3782        3783       GitHub  https://github.com/torvalds/linux  torvalds   \n",
      "\n",
      "       repo                                         subject status  \\\n",
      "3778  linux                                         MD pull   open   \n",
      "3779  linux  Auzeuner DVB-T2 (Astrometa 2018) driver update   open   \n",
      "3780  linux  Auzeuner DVB-T2 (Astrometa 2018) driver update   open   \n",
      "3781  linux  Auzeuner DVB-T2 (Astrometa 2018) driver update   open   \n",
      "3782  linux  Auzeuner DVB-T2 (Astrometa 2018) driver update   open   \n",
      "\n",
      "          category  pull_request_number                date  pull_request_id  \\\n",
      "3778  pull_request                  568 2018-07-05 18:34:52        338681544   \n",
      "3779  pull_request                  567 2018-07-05 04:41:44        338425270   \n",
      "3780  pull_request                  567 2018-07-05 04:44:51        338425270   \n",
      "3781  pull_request                  567 2018-07-06 13:09:44        338425270   \n",
      "3782  pull_request                  567 2018-08-23 21:24:18        338425270   \n",
      "\n",
      "             user                                               body  \n",
      "3778  KernelPRBot  Hi @wburnett01!\\n\\nThanks for your contributio...  \n",
      "3779     kapitanf  Hi Linus,\\r\\n   here is rtl28xxu.c rtl28xxu.h ...  \n",
      "3780  KernelPRBot  Hi @kapitanf!\\n\\nThanks for your contribution ...  \n",
      "3781     kapitanf    Very helpful, many Thanks.\\r\\n\\r\\nRegards Filip  \n",
      "3782  negrumanuel  Hi Filip, \\r\\n\\r\\nDid you pursue this any furt...  \n"
     ]
    }
   ],
   "source": [
    "if(\"notebooks\" in os.getcwd()):\n",
    "    os.chdir(\"../SentiCR\")\n",
    "    from SentiCR.SentiCR import SentiCR\n",
    "\n",
    "sentiment_analyzer = SentiCR()\n",
    "\n",
    "if(\"notebooks\" not in os.getcwd()):\n",
    "    os.chdir(\"../../notebooks\")\n",
    "augurApp = augur.Application('../augur.config.json')\n",
    "connect,list1,path = augurApp.github_issues()\n",
    "DB_STR = 'mysql+pymysql://{}:{}@{}:{}/{}?charset=utf8'.format(\n",
    "            list1[0], list1[1], list1[2],\\\n",
    "            list1[3], list1[4]\n",
    "        )\n",
    "db = s.create_engine(DB_STR)\n",
    "\n",
    "table_names = s.inspect(db).get_table_names()\n",
    "print(table_names)\n",
    "\n",
    "if(\"github_pull_requests_2\" in table_names):\n",
    "    lists_createdSQL = s.sql.text(\"\"\"SELECT repo FROM github_pull_requests_2\"\"\")\n",
    "    df1 = pd.read_sql(lists_createdSQL, db)\n",
    "    #print(df1)\n",
    "    val = True\n",
    "\n",
    "col = 'score','sentiment'\n",
    "\n",
    "SQL = \"\"\"SELECT * FROM github_pull_requests_2\"\"\"\n",
    "df_users =  pd.read_sql(SQL, db)\n",
    "print(df_users.head())\n",
    "print(df_users.tail())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "713\n"
     ]
    }
   ],
   "source": [
    "grouped = df_users.groupby('subject').groups\n",
    "num = 0\n",
    "col = 'score','sentiment'\n",
    "columns1 = 'augurmsgID',\"backend_name\",'repo_link',\"owner\",\"repo\",\"subject\",\\\n",
    "          \"status\",\"category\",\"pull_request_number\",\"date\",\\\n",
    "          \"pull_request_id\",\"user\",\"body\"\n",
    "df3 = pd.DataFrame(columns=col)\n",
    "df_list = pd.DataFrame(columns = columns1)\n",
    "df_list = df_list.append(df_users)\n",
    "for group in grouped:\n",
    "    #print(group)\n",
    "    issue_num = (df_users.loc[df_users['subject'] == group]['pull_request_number']).values\n",
    "    messages = (df_users.loc[df_users['subject'] == group]['body']).tolist()\n",
    "    id1 = (df_users.loc[df_users['subject'] == group]['body']).tolist()\n",
    "    #print(\"Issue number: \",issue_num[0])\n",
    "    #break\n",
    "    for i in messages:\n",
    "        #print(i)\n",
    "        if(i == None):\n",
    "            i = \" \"\n",
    "        score=sentiment_analyzer.get_sentiment_polarity(i)\n",
    "        score = int(score[0])\n",
    "        compound = parts = 0\n",
    "        sentiment = \"Positive\"\n",
    "        if(score == -1):\n",
    "            sentiment = \"Negative\"\n",
    "        li = [ [score,sentiment]]\n",
    "        df_temp = pd.DataFrame(li,columns=col)\n",
    "        df3 = df3.append(df_temp)\n",
    "        #if(avg < -0.5):\n",
    "            #print(i,avg)\n",
    "            #print(\"\\n\\n\\n\\n\\n\\n\")\n",
    "    #if(num == 40):\n",
    "    #    break\n",
    "    num+=1\n",
    "    #print(\"\\n\\n\\n\")\n",
    "print(num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  augurmsgID backend_name                      repo_link   owner repo  \\\n",
      "0          1       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "1          2       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "2          3       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "3          4       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "4          5       GitHub  https://github.com/nodejs/CTC  nodejs  CTC   \n",
      "\n",
      "                subject  status      category pull_request_number  \\\n",
      "0  add meeting schedule  closed  pull_request                  18   \n",
      "1  add meeting schedule  closed  pull_request                  18   \n",
      "2  add meeting schedule  closed  pull_request                  18   \n",
      "3  add meeting schedule  closed  pull_request                  18   \n",
      "4  add meeting schedule  closed  pull_request                  18   \n",
      "\n",
      "                 date pull_request_id          user  \\\n",
      "0 2016-10-06 04:03:18       181319945         Trott   \n",
      "1 2016-10-06 13:17:53       181319945      mhdawson   \n",
      "2 2016-10-06 21:16:54       181319945  williamkapke   \n",
      "3 2016-10-06 21:31:31       181319945       jasnell   \n",
      "4 2016-10-06 21:34:01       181319945   MylesBorins   \n",
      "\n",
      "                                                body score sentiment  \n",
      "0  Not sure that the README is the best place for...     0  Positive  \n",
      "1                                             LGTM\\n     0  Positive  \n",
      "2  This is a bit of a tangent but.. should the Me...    -1  Negative  \n",
      "3  I would definitely prefer the meeting notices ...     0  Positive  \n",
      "4                       +1 on meeting notices here\\n     0  Positive  \n"
     ]
    }
   ],
   "source": [
    "#print(df3)\n",
    "#print(df_list)\n",
    "df3 = df3.reset_index(drop=True)\n",
    "#print(df3.head())\n",
    "df_list = df_list.reset_index(drop=True)\n",
    "combine = (df_list.join(df3))\n",
    "\n",
    "print(combine.head())\n",
    "\n",
    "combine.to_sql(name='github_pull_requests_sentiment_scores_senticr',\\\n",
    "               con=db,if_exists='replace',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(47,)\n",
      "(47,)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEUCAYAAADXzmpaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XmYXFW57/HvL0MzxgBJGIKEIAnKaCANigNylAgisyIIB0FFUI5wvSoKCIqKDOLxOiBHBj0MCogoEBGZojIjSTBAkCkCMoaEEBAQzPTeP9bqpFKp7t7VXbuq0/37PE89XXt819q1e797rz0pIjAzMytiUKsLYGZmKw8nDTMzK8xJw8zMCnPSMDOzwpw0zMysMCcNMzMrzEmjD5E0VlJIGpK7/yzp8FaXqy+S9ISkXVpdjt6SNEbSq5IGdzFOSBrXw/kPqHWonvrm5f6WssvU3zhplCBv0F7PK+Xzkv5X0poNjnGypIU5xkuS7pC0YyNjNFqzNvSSdpb0dFW/kyX9osSYbZJe6Pidq7s7ExFPRsSaEbE4TzcgNvKSRlf/Rs2Wl/tjRcaVdIKkU/P3N0n6gaQn8//frNw9Mnd3fJZUbAdelXRwnn4zSb/O68fLku6T9MWudhz6EieN8uwZEWsC2wHbAyeWEONXOcZI4E/Ar0uIYcXsBMyIiFc76bbl7Q5c1+pC1GF34FpJbcAUYEtgN+BNwLuAecAOORGtmf8vnyRvB/Lnl5I2Bf4CPAVsHRHDgf2BdmBY86tVPyeNkkXEM8AfgK1gxb3tRuwBR8Qi4JfAhpJGVcx7D0kzKo5EtqkYtq2keyS9IulXki6TdEoedpik2ypjVDaRSFpF0vfyntbzkn4qabU8bKSka3LMFyXdKmmQpIuBMcDv8l7XVyStKukXkubl8adKWq+Oqk/Ie2kv5zqsKmkN0vIeXbGHdxBwAnBA7r43l/XPkk6TdHeex9WS1snD6i3b7sC1tbolbSLplrysb5L0k47fXBVNkpK+A7wXOCuX86yK+e0i6VFJ8/P0ytMvt/6oqokz27RWHWuR9BlJD+ay/k3Sdrn/5nl5vSTpAUl7VUxzgaSzJf0hl/t2Sevnve/5kh6StG1nyyv/T3y5+resKtOsvD5NljS6YtikPP+X8/JSVX0+leszX9L1kjauGFa5Tu+e6/uKpGckfblivLWBzYA7gU+Q1uN9I+JvEbEkIuZExLcjovL378w3gTsi4osR8RxARDwcEQdFxEsFpm85J42SSdqI9A/y1xJjtJFW5nnA/NxvO+DnwJHACOAcYHLe4LcBVwEXA+uQjlA+UkfIM0j/RBOAccCGwNfzsC8BTwOjgPVIG+uIiENYfs/ru8ChwHBgo1zGzwKv5/IfJ+mabsrxMdLe3ibANsBhEfEa8CHg2Yo9vEuAU8lHZhHx9op5fAL4FDAaWAT8KPevt2y7A7/vpPsS4O48n5OBQ2pVJiK+BtwKfD6X8/MVg/cgHbG+Pdd7164XzXI6q+NyJO2fy/cJ0h70XsA8SUOB3wE3AOsCRwO/lPTWisk/RjqaHgn8m7SBvSd3XwF8vyLOUNKR2I1V0y/3W+Zx3w+clodvAPwDuCwPGwn8piLu34F3V8TZh7T+7UdaH28FLu1kGf0MODIihpF28P5YMWxXYEpuQtwFuK4XR5C7kJbHSstJozxXSXoJuA24mbTRarSP5RivA58BPpqPOsjd50TEXyJicURcSPpnfmf+DAV+EBELI+IKYGqRgHkP9zPA/42IFyPiFVLdDsyjLCT9c2+c531rdP6As4WkDem4XMbpEfFPgIg4PSL26KY4P4qIZyPiRdJGbUKROlS5OCJm5mRzEmmZDq6nbEonU4dGxMPV3ZLGkDb2X4+IBRFxGzC5B+U8PSJeiognSU2R9dS1szpWOxz4bkRMjWRWRPyDtL6smcuwICL+CFwDfLxi2ivzMnoDuBJ4IyIuyhvaXwGVRxo7AffmdadDZ7/lwcDPI+KeiPg3cDywo6SxpMT8t4i4IiIWAj8AZlfM80jgtIh4MP9fnEo6Ot2YFS0EtpD0poiYHxH3VAz7MMuOIkcAz9WYvqjeTt9yThrl2Sci1oqIjSPiqIh4vYQYl0fEWqQ9+pnAxIphGwNfys0JL+XkshFpb3M08EzVxvwfBWOOAlYHplfM97rcH+BMYBZwg6THJB3XxbwuBq4HLpP0rKTv5r3Qoio3EP8ibdjq9VTF93+QkunIOstWuVGp7h4NvBgR/+okZlG9qWtnday2EWlvvdpo4KmIWFI1nw0rup+v+P56je7K8lY35UHn9RtNxbqZ9/Dn5dijqahbXp8r67ox8MOK9fRFUvNVZbk7fCSX6x+Sbla+qETSIGASy86/zCPtFPVUb6dvOSeN5nuNtNHtsH5vZxgRL5D2qk6W1LFCPgV8Jyeujs/qEXEpaU9nw4528WxMZ2WUVFnGF0gbgS0r5js8n/gjIl6JiC9FxFuAPYEvSvpAR1Gryr0wIr4ZEVuQTibuQWoa6a1aRzadHe1sVPF9DGmP84U6y9ZV09RzwDqSKn/zyphFy9mZIutTzTrWGO8pYNMa/Z8FNsob0Mr5PFNfUZeqXl5deZa08QdA6ZzViBz7OSrqltfnyro+RWpyqvwfWC0i7qgOko+u9iY1v10FXJ4HbQ88ERFzc/dNwK65HD1xE/U1Bfc5ThrNNwM4UNJQSe3ARxsx04h4iLRn/JXc6zzgs5LeoWQNSR+WNIzU3rwIOEbpBOx+wA4Vs7sX2FLShHxC8uSKOEvyvP+fpHUBJG0oadf8fQ9J4/I/8D+BxfkDae9z6XXxkv5D0ta5qeSfpI1Zx7i98TwwQtLwqn5jqzZ8AP8paYu8Uf8WcEVELC5aNqULAHYA/lyrOzfvTCMl9La8B7tnN2Wv596BGcBOSvd7DCc131SrWcca450PfFnSxLzOjMtNOX8hJaev5PV251yHy+ooJ5AuCgBWyetrEZcAn8zr4iqkJqa/RMQTpMSzpaT9lE78H8PySfOnwPGStsyxh+fzNtVlapN0sKThuZmrY72FFY8iLyYlo99IepvSRR4jlC7J3b1Afb4BvEvSmR07Y3k5/0LSWgWXSUs5aTTfSaS9ufmkKykuaeC8zwSOkLRuREwjnXs4K8eaRT65GBELSCcHD8vDDgB+2zGTiHiEtHG5CXiUdF6m0lfz/O6S9M88XsdJ0fG5+1VScjo7Iv6ch50GnJibC75M+ge/gvRP+iDp3E/HVUUnSPpDTxZC3iBdCjyWY41m2eXI8yRVtldfDFxAah5ZlbThoY6yfQC4M7fl1+qG1C6/I6lp4hRSG/+/Oyn+D4GPKl3tU/OEdVVdb8zzuw+YTjrXUK2zOnbc4PbePK9fA98hrZOvkPa418nry16kCwxeAM4GPlHHhr9S9Ua4SxExhfQ/8xvSkcWm5PNn+Qh7f+B00rIdD9xeMe2VpIs2Lsvr6cxch1oOAZ7I430W+M/cf7mmtHxeZRfgIdKJ/H+SLnIYSUqu3dXn76R1YSzwgKSXc92mkZZ5n6fOz1HaQCLpAuDpiCjjfpI+SdKfgV9ExPm9mMfZwMyIOLtWdyfT/Ap4KCK+0dO4KytJ1wJnRbHLU8ssxyDS0cTG+eKCWuOsRzqSG93FxRwDjo80zHpnBulqoc66kbS9pE1zU8ZuwN6kvfiB6M+kq79abSvgDZY/AV9tOPBFJ4zlDel+FDPrTESc21V3tj6p+W8E6R6Wz0VEafft9GWR7s9pKUkfAc4Fvpqb3mrKzbSPNK1gKwk3T5mZWWFunjIzs8L6XfPUyJEjY+zYsa0uhpnZSmX69OkvRMSo7sbrd0lj7NixTJs2rdXFMDNbqUgq9FQIN0+ZmVlhThpmZlaYk4aZmRXmpGFmZoW19ER4vjv2h8Bg4PyIOL1q+CrARaRHfs8DDsgPKmu4O390GNvPu5rBLGExg5g6Ym92POaCMkItNXXyOWw+/STWqHgM0WuswoMTv832ex25dJyN7jmTdWMuczSKp7Y7lu33OrLT/r0pS0/mV2u6RU/c2emy7EmcVkyzhEEMZgnPaxSPr/MeNnnxtk7nVSsWwLh7vsVa+V09L2kYs7Y7aelvV2sY0GmZG/X7zBm6IVssuK/u9fy+77yPrRfMWNp9f9sEtvnazYVjdzxOuWjMRq7ffeV/peg8X9YwIBger9ZcDzpbr5qlZTf35aeHPkJ6Vv3TpJcAfTwi/lYxzlHANhHxWUkHkl6xeEBX821vb496r56680eH8c55V1L5oPAIuGvEvqUljqmTz2HC9K8yVCsu/0Uh/jrxDAC2mn4iq2nZTauvRxszRnyYCfN+v0L/mRNP6dHKM3XyOTXjdDe/WtMtirSxrbUsh4zdse44PSlbo6apLH9lfSrnVWu6BTEEsXiF33ZBDGH6iD3Zbt5kVtHyD5ldFGIJg2nToqX9OuJA7fWgJ79PdV2KrOcdCaN6uq4SR3fLs6uYPV0fy55XGfPrbJ6VKteDbaYfv8K6syCGcO/EU3udOCRNj4j2bsdrYdLYETg5IjoeqX08QEScVjHO9XmcO/Ojj2cDo7p6FkxPksaib6zNEC1ZsX8MYsg359c1r6JmnzyO9Znb+fD8TqNa4yyKQTXLO5tRrH/yrIaVpbv5dVeHSotiEC9oRN1xelK2Rk7TmY551TtdZ79dV3Gg9nrQqN+nu/U8vjF8uYSxtH+Avvlyj2J3FbOn62PZ8ypjfl3Ns3r+UHs96G38DkWTRiubpzZk+bdsPQ28o7NxImJRfozwCKpeICPpCOAIgDFjxlCvwdT+J+6sfyOsG3Ohxj/isuEvAFFznM7KlaZpXFm6m193dag0mCU9itPqaTof/4UeTVfvOtXVetCo36eM9by72F3F7On6WPa8yphfV/Nccf6114Pexq9XK0+E16p+9RFEkXGIiHMjoj0i2keN6vaGxhUs7mQxdNa/Eeao63LO0chOx+msXHNU6w2ePS9Ld/Prrg6VFjOoR3FaPU3n44/s0XT1rlNdrQeN+n3KWM+7i91VzJ7Wt+x5lTG/ruZZPf+uxutN/Hq1Mmk8zfKvZnwz6dWONcfJzVPDSe/5baipI/amusErIvUvy1PbHcvCqL3bsCjEU9sdy1PbHcvr0bbcsNejjakj9q7Zv+MEbE/K0pP51ZpuUQzqdFn2JE4rp6ksf2fzqjXdghhS87ddEEOYOmJv/h2DVxi2KMSCWP7AvyNOI3+fnqzn97dNqDnd/W0T6opdNGZP61v2vMqYX2fzrDX/p7Y7tua6syCG9Cp+vVqZNKYC4yVtIqmN9DauyVXjTAYOzd8/CvyxjGfb73jMBdw1Yt+lG7xFMajUk+AA2+91JDMmnsGrsQoRLP28Gqvw14lnsP1eR7L9Xkcyc+IpzGYUS0LMZhQzJ57CjsdcULN/T0+EdRanu/nVmu6vE0/vdFn2JE5rplmW/GYzirtG7NvpvGrFunfiqcyYeAbzWXPp7zqfYdw78VR2POYC7pt42grD/jrxDO6deGrNOI38fe5vm1D3er7N125emjg6Pt1dPVW9PDumKxKzp/Ute15lzK/WPOczjPmsWXM9qLXuNOIkeD1a+mj0/E7dH5Auuf15RHxH0reAaRExOb+f+mJgW9IRxoER8VhX8+zJiXAzs4FuZTgRTn7l47VV/b5e8f0N0juAzcysD/Ad4WZmVpiThpmZFeakYWZmhTlpmJlZYU4aZmZWmJOGmZkV5qRhZmaFOWmYmVlhThpmZlaYk4aZmRXW0seI9DVbX7h1Q+az/hrrM26tcWw6fFM2XWvT9H2tTVl96OoNmb+ZWas4aZRg9muzmf3abG575ramxDvwrQfy+W0/z/BVhjclnpkNXC19ym0ZmvWU24hg9muzefSlR3nspcf4+8t/Z9b8Wcx6aRZvLH6j9PjNdOgWh/KFiV9gyCDvY5j1V33+HeFl6S+PRn9t4Wtc+/i1XP7w5Tz04kOtLk6pjt/heA7a/KBWF8NsQHPSsC49Ov9RTrr9JB6Y90Cri1K6yz58GVuO3LLVxTDr05w0rKWunnU1J95+YquLUboxw8Zw0YcuYsRqI1pdFLNecdKwfmvh4oV8665vcdWsq1pdlFINaxvGse3Hsve4vRkkXx1v5XLSMOuhuf+ayxf+9AXue+G+VhelVDtusCNHb3s0W49qzKXmtnJz0jDro2a+MJMTbjuBx19+vNVFKdXOb96Zr+zwFTYatlGri2IFOGmYDVAz5szgggcuYMqTU1pdlNK0DWrjcxM+x0FvO8g3zTaIk4aZNdzCxQu5/JHLOf3u01tdlFJtOnxTjppwFLtsvMuAOZ/kpGFmK51XFrzCNY9dw6UPXdpvm+82X2dzPjj2g+w6dtc+1XTnpGFmVuXxlx/nvPvO43eP/a7VRWmIj4z/COusus7Sz3+M+Q9WG7Jaj+blpGFm1kQRwayXZnHdE9dx3ePX8eQrTza9DJM2nsT3d/5+j6YtmjT8MCEzswaQxPi1xzN+7fEcve3RDZ//oiWLeHj+w9z93N088+ozvHnNN7P60NV58Y0Xl36+OPGLDY9bzUnDzGwlMGTQELYcsSVbjmjtI3EGxmUBZmbWEC1JGpLWkXSjpEfz37VrjDNB0p2SHpB0n6QDWlFWMzNbplVHGscBUyJiPDAld1f7F/CJiNgS2A34gaS1mlhGMzOr0qqksTdwYf5+IbBP9QgR8UhEPJq/PwvMAUY1rYRmZraCViWN9SLiOYD8d92uRpa0A9AG/L2T4UdImiZp2ty5cxteWDMzS0q7ekrSTcD6NQZ9rc75bABcDBwaEUtqjRMR5wLnQrpPo86implZQaUljYjYpbNhkp6XtEFEPJeTwpxOxnsT8HvgxIi4q6SimplZQa1qnpoMHJq/HwpcXT2CpDbgSuCiiPh1E8tmZmadaFXSOB2YJOlRYFLuRlK7pPPzOB8DdgIOkzQjfya0prhmZgZ+9pSZmVH82VO+I9zMzApz0jAzs8KcNMzMrDAnDTMzK8xJw8zMCnPSMDOzwpw0zMysMCcNMzMrzEnDzMwKc9IwM7PCnDTMzKwwJw0zMyvMScPMzApz0jAzs8KcNMzMrDAnDTMzK8xJw8zMCnPSMDOzwpw0zMysMCcNMzMrzEnDzMwKc9IwM7PCnDTMzKwwJw0zMyvMScPMzApz0jAzs8JakjQkrSPpRkmP5r9rdzHumyQ9I+msZpbRzMxW1KojjeOAKRExHpiSuzvzbeDmppTKzMy61KqksTdwYf5+IbBPrZEkTQTWA25oUrnMzKwLrUoa60XEcwD577rVI0gaBPw3cGyTy2ZmZp0YUtaMJd0ErF9j0NcKzuIo4NqIeEpSd7GOAI4AGDNmTD3FNDOzOpSWNCJil86GSXpe0gYR8ZykDYA5NUbbEXivpKOANYE2Sa9GxArnPyLiXOBcgPb29mhMDczMrFppSaMbk4FDgdPz36urR4iIgzu+SzoMaK+VMMzMrHladU7jdGCSpEeBSbkbSe2Szm9RmczMrBuK6F+tOe3t7TFt2rRWF8PMbKUiaXpEtHc3nu8INzOzwgonDUnvkfTJ/H2UpE3KK5aZmfVFhZKGpG8AXwWOz72GAr8oq1BmZtY3FT3S2BfYC3gNICKeBYaVVSgzM+ubiiaNBZHOmAeApDXKK5KZmfVVRZPG5ZLOAdaS9BngJuC88oplZmZ9UaGb+yLie5ImAf8E3gp8PSJuLLVkZmbW53SbNCQNBq7PjwVxojAzG8C6bZ6KiMXAvyQNb0J5zMysDyv67Kk3gPsl3Ui+ggogIo4ppVRmZtYnFU0av88fMzMbwIqeCL9QUhuwWe71cEQsLK9YZmbWFxVKGpJ2Jr2W9QlAwEaSDo2IW8ormpmZ9TVFm6f+G/hgRDwMIGkz4FJgYlkFMzOzvqfozX1DOxIGQEQ8Qnr+lJmZDSBFjzSmSfoZcHHuPhiYXk6RzMysryqaND4H/BdwDOmcxi3A2WUVyszM+qaiSWMI8MOI+D4svUt8ldJKZWZmfVLRcxpTgNUqulcjPbTQzMwGkKJJY9WIeLWjI39fvZwimZlZX1U0abwmabuODkntwOvlFMnMzPqqouc0vgD8WtKzpBcxjQYOKK1UZmbWJ3V5pCFpe0nrR8RU4G3Ar4BFwHXA400on5mZ9SHdNU+dAyzI33cETgB+AswHzi2xXGZm1gd11zw1OCJezN8PAM6NiN8Av5E0o9yimZlZX9PdkcZgSR2J5QPAHyuGFT0fYmZm/UR3G/5LgZslvUC6WupWAEnjgJdLLpuZmfUxXSaNiPiOpCnABsANERF50CDg6J4GlbQO6aT6WNLj1j8WEfNrjDcGOB/YiHTV1u4R8URP45qZWe8UeUf4XRFxZURUvub1kYi4pxdxjwOmRMR40t3mx3Uy3kXAmRGxObADMKcXMc3MrJeK3tzXaHuTXupE/rtP9QiStgCGRMSNkO5Cj4h/Na+IZmZWrVVJY72IeA4g/123xjibAS9J+q2kv0o6Mz8ocQWSjpA0TdK0uXPnllhsM7OBrbQroCTdBKxfY9DXCs5iCPBeYFvgSdI5kMOAn1WPGBHnku8baW9vj+rhZmbWGKUljYjYpbNhkp6XtEFEPCdpA2qfq3ga+GtEPJanuQp4JzWShpmZNUermqcmA4fm74cCV9cYZyqwtqRRufv9wN+aUDYzM+tEq5LG6cAkSY8Ck3I3ktolnQ8QEYuBLwNTJN1PemPgeS0qr5mZ0aK7uiNiHukO8+r+04DDK7pvBLZpYtHMzKwLrTrSMDOzlZCThpmZFeakYWZmhTlpmJlZYU4aZmZWmJOGmZkV5qRhZmaFOWmYmVlhThpmZlaYk4aZmRXmpGFmZoU5aZiZWWFOGmZmVpiThpmZFeakYWZmhTlpmJlZYU4aZmZWmJOGmZkV5qRhZmaFOWmYmVlhThpmZlaYk4aZmRXmpGFmZoU5aZiZWWFOGmZmVpiThpmZFeakYWZmhbUkaUhaR9KNkh7Nf9fuZLzvSnpA0oOSfiRJzS6rmZkt06ojjeOAKRExHpiSu5cj6V3Au4FtgK2A7YH3NbOQZma2vFYljb2BC/P3C4F9aowTwKpAG7AKMBR4vimlMzOzmlqVNNaLiOcA8t91q0eIiDuBPwHP5c/1EfFgrZlJOkLSNEnT5s6dW2KxzcwGtiFlzVjSTcD6NQZ9reD044DNgTfnXjdK2ikibqkeNyLOBc4FaG9vj56V2MzMulNa0oiIXTobJul5SRtExHOSNgDm1BhtX+CuiHg1T/MH4J3ACknDzMyao1XNU5OBQ/P3Q4Gra4zzJPA+SUMkDSWdBK/ZPGVmZs3RqqRxOjBJ0qPApNyNpHZJ5+dxrgD+DtwP3AvcGxG/a0VhzcwsKa15qisRMQ/4QI3+04DD8/fFwJFNLpqZmXXBd4SbmVlhThpmZlaYk4aZmRXmpGFmZoU5aZiZWWFOGmZmVpiThpmZFeakYWZmhTlpmJlZYU4aZmZWmJOGmZkV5qRhZmaFOWmYmVlhThpmZlaYk4aZmRXmpGFmZoU5aZiZWWFOGmZmVpiThpmZFeakYWZmhTlpmJlZYU4aZmZWmJOGmZkV5qRhZmaFOWmYmVlhThpmZlZYS5KGpP0lPSBpiaT2LsbbTdLDkmZJOq6ZZVzZTJ18DrNPHseSbwxn9snjmDr5nB6NszLpb/VpFi+3xukry7KZ5WjVkcZMYD/gls5GkDQY+AnwIWAL4OOStmhO8VYuUyefw1bTT2R95jJIsD5z2Wr6icutOEXGWZn0t/o0i5db4/SVZdnscrQkaUTEgxHxcDej7QDMiojHImIBcBmwd/mlW/lsdM+ZrKYFy/VbTQvY6J4z6xpnZdLf6tMsXm6N01eWZbPL0ZfPaWwIPFXR/XTutwJJR0iaJmna3Llzm1K4vmTdqF3ndeOFusZZmfS3+jSLl1vj9JVl2exylJY0JN0kaWaNT9GjBdXoF7VGjIhzI6I9ItpHjRrV80KvpOaodp3naGRd46xM+lt9msXLrXH6yrJsdjlKSxoRsUtEbFXjc3XBWTwNbFTR/Wbg2caXdOX31HbH8nq0Ldfv9Wjjqe2OrWuclUl/q0+zeLk1Tl9Zls0uR19unpoKjJe0iaQ24EBgcovL1Cdtv9eRzJx4CrMZxZIQsxnFzImnsP1eR9Y1zsqkv9WnWbzcGqevLMtml0MRNVt8SiVpX+DHwCjgJWBGROwqaTRwfkTsnsfbHfgBMBj4eUR8p7t5t7e3x7Rp08orvJlZPyRpekR0egtEhyHNKEy1iLgSuLJG/2eB3Su6rwWubWLRzMysC325ecrMzPoYJw0zMyvMScPMzApryYnwMkmaC/yj4OgjgWbf1dTsmP09XjNj9ue6tSpes2P29/r1JubGEdHtjW79LmnUQ9K0IlcLrMwx+3u8Zsbsz3VrVbxmx+zv9WtGTDdPmZlZYU4aZmZW2EBPGucOgJj9PV4zY/bnurUqXrNj9vf6lR5zQJ/TMDOz+gz0Iw0zM6uDk4aZmRXW75OGpFrv5eg3+nv9OjSrnq1YngPlN7TGatV60++TBrBqx5d++s/Z1v0o/UKzfrulD/Fs4vqyVo5X+gNEJb1VUlP/7yW9X9L6TYx3kKS35+/N2tlYq+J7s9ablmy/+23SkPRBSXcAZ0k6GCBKPOsvaR9JP5a0TlkxquLtLuk64IeSDmlSzH0kfbsZsSpi7i7pauBMSTuXGGc3SdcD38uP7i91fckxh0u6Abgux1tUYqxJkv4CHE6T/u8lvUvSA8BhwJpNiLeLpFtJr1PYFpryG35I0s3ATyQd36SYH5Z0DfBtSe8uM1ZNEdHvPqT3dNwF7AN8gPTypuPzsEENjiVgP+BB0tsGP9LoGFXxhgAnkF5StRtwMHApsGdJ8UR6n8nhwCxgIfDekn8/kY6g/hu4G/gQ8HXgHOAdJcT5HnA7sBdwFHAJML4J6+kqwK+A14D9c7/BDa7fUOBbwKPAftXDS6zbYOB/gYOasK6sBlwO/BnYFTgbOKrRy7NG7B2Av+T15n3Ar4GtSq7vxPw/sTvpxXTnAoflYaVtdyo//e5IIx8argfcGxFXRcQU4Djgy5JGRsSSRh4+Rvq1HgPeA/wf4D9Jr6YtRaTAqqWBAAAL/0lEQVS90ceAAyPiOlJCfJaSmqkiWUxKGNuSNqqlHm3kmAuAR0gbnT8A55OacRaXEOc64H0RMRm4g5QYH29UnFokDSbV5y7gANJLyYiIxY1aP3P9FgJLgCsi4rc59nslDW1EjC68ibRBv1ZSm6RDJI3Lb+FsWBNOruPrwC8jYueIuJ70Gx6Shzdsfanh3cAteb15irRu/r2j+a+kZqpdgFsjvWvoamA2cLSk4Y3etnWmXyQNSYdKmgRLN+KvAu/qaCqKiL+R9gJ+3Oh42cyImBcRvyFtcPbr+OcoKd5vgcclDY2IV0hJavVGxcsxj5F0nqTDc6+bI+KViDgPWEPSp/N4DVuHKmJ+Jvc6D3hMUlukF3QNA0Y0MM7hABFxU0Qsym+K/C2wGXCqpAPy+L3+R6yI+SlJyhuzfwIfjohrgPskfV3SVhERvYlZEeuI3OunwAaS/lfS/cBXgJ8Bn8rjN7J+n869BgFvAbYh/e/tCZxKOlqEXp6jql5XIuLq3H8wKeE/IGmj3sToLiZwE3CQpB8DtwCjgf8BvllizD8Be0haOyfLhaT16CtQftMYOchK+wHWBq4AngPuo+JQFLgIuLhq3L8AmzQ6HukfpONGyXcDU4Dtqqatuymgq3gV46wKXAW8tYHL9TDSHvBuwM3A8cCmFcM/BDwArF1yzHFVy2IKsH6D45zQEYfU3LBZ/r47cD0wtoS6nQBsCqwLnJLH+RSwCJiWu4c2KNaJedntA/wSeBtpg7038HtgTAn1O5HUZHQ68HfggDzemsBcoL2E5fmWiuFbk5pvh5W4fp5EOlJcG/g+uXkY2ByYCWxZQsyv5XXmx8A1wK2kJsBdSclqjUbVt6vPSn2kERHzgRtIP9R0Urt3h88Du0naPne/BtwLLCgjXuRfOSJuB2YAH5L0to69vY7hjYpXYS1g1Yh4WNJGkj5Sb5waPgCcEan560ukxHRwRbn+QDqHc4SkYZL2LynmQRXDxwIvR8RsSW+W9P4GxWkj1y0i7o6IR/J4fyNt4BpxcrpW3fYHXietJzcAxwB/ZNlj/XsatzrWKsCREXEVcEREPJTXxfuAl0h7qr1Vq35HkdbXNfKHiHgVuIy0oW1kvDZSszA5zv2kZXtgL+N0FXMo8Pn8P7oZy363h4A7Scu90TFXBT4REUeTlu+3IuKTwBukbcBrDYjZrZU2aVQcUl8UES+RTn7tJ2ljgIj4J+kw8SRJh5L2frYiNV01NF6ktsTBFU01PyDtKd9M2jOouwmgQLyOyzPfAgyT9AXS+Y1un4ffRcyO8v8V2AMgIqaR/glGV12p8VXgNNIJ1h5fTlkg5nvz8A2BwZKOJu0h1xWzizh3kZpuqq9COYzU5DevnjgFY94BbEI6D3YjcHdETIiIDwI7S9qk3p2MLmLdDmwi6d1VG5VDSUcD83tUua5j3gZsAWxAajbZTdKekk4kHYk/2OB4d1Gxfub/nRuAVXvb9NbNbzhW0hakZH++pNVZtp15uoSYtwPjJb0nIp6MiBvzeB8mHdE1xUqTNCRtKWnpPRcVe/Zv5L9TgT8A36kY5yzSBnwisDHw0Yh4uYx4EbE4b8zXA84irUgTIuKUyukbGK9jT3QisCMwjtQ+/tMi9csx3y1p04qYS/LX24FBknbK3TNJTWSj83TjSEnsKlIzXOFzRT2I2ZEcJpHaxccBu0fEJSXV7ROSZpI26p+L1G7c6Lo9ADxDOkfz9Yg4sWI2YyKi25Pwddbv2Yr6fUTSvaSdjc91rF8Nrt9M0kZzYkRcRDqn8h5gDLBHRBTaoPb0N8z/O+sCr/Ug+dZbx7dFxPeBh0lNyVuQrlKbU1LMZ0nJGEk7KV3uO560jJujGW1gvfmQTqTdBlwJbFjRX1RdYkZaKe8CtiRdQdXRVl34srtexBtF2tAMpo524l7WbwRpr6auS2CB7Uh7Yv+m4txLRzxgHeDLwE9Ydh7lf4Bj8/dR1HlJai9iHpe/7wR8oMQ4X8nfJwDvakLdflqxPAdX/9Yl1m9rYMcm/XZfrR63Getn7m5rUh07LuVvA9ZqUsyO33IMDTh3Uu9nZTjSOJF0ueC+EfEMpCskIlkiaTVJawJExJOkje/9pKahN+X+9Vx219N4t5JODC/O/cuOdwvp9YwzI+LWIoEkDZV0Duna7h+RTvTuXBGzYw/nlVyfNtLNbkNJ7dDzcjnmRsSjTYo5J8e8JdLl02XFeSHHmRERdzShbmuxbHkurhi37PrdHxF3NqF+a5POCZHjdlm/BsVb2pQY6VLqZtTx+Y54kZqRmxGz47d8MiIeKBKzoZqdperIwoNIV5j8vKLfJNI/25Dc/W3SJZLb5O6Pk05IfZc6rz7p7/Hy9GuSTvqulrsPI13WOqRinG+SLpF8G+kw+ALS0c059OBGqWbF7M91Gwj1Gyh1bFXMRn5aFriThflO8iWPuXsY6UTrHqT28+tJl9IeT7qa5hKWvyzzndRxSW1/j1cdk6rLfoFPAz/tGEZqKruE5S+vHUSdly42K2Z/rttAqN9AqWOrYpb1aXkB8gJZi3RFzCuk5po1KoadANwD7JW7dyLdCbljxTj17tH063hdxaTiXAnppPLz5PstKldm6miDbnbM/ly3gVC/gVLHVsUs+9NXzmmsQdrLPjp/36li2DWkve6OBwFOI906/waky9Oi/kcF9Pd4ncaMZEm+rO+JPM77OoZVxOy2DbqFMftz3QZC/QZKHVsVs1QtSxr50sb3SXpTpBPA55IeOvYG8A5JGwJExH3AscB/SRpJuolna5adRCy0UPt7vIIxOy67VJ5vxyW+HQlKfTVmf67bQKjfQKljq2I2U7Ofqy9JG0j6E+nmooOB/1F6kOAbEfEv0vNc1gaW3vEbET8jPcn1ZNJTZA+PAlco9fd4PY0ZEaF0lcarpMPkd3b070sx+3PdBkL9BkodWxWzZaJJ7WAsu854M+AX+fsQ0nNUfls17v8FTgGGU3HyhzquGOrv8XoZc/W+HrM/120g1G+g1LFVMVv5KT9AWninAmeQ2uz2BC6sGC7S3Zzvq+i3JulO7rtJJ4hGO97AiNmf6zYQ6jdQ6tiqmH3hU+7M04K8l3QX42dIN6TtBjwJ7FAx3ueAP1V0H0B6sOB5wLqONzBi9ue6DYT6DZQ6tipmX/mUO3N4L3BIRffZeSEeBkzP/QaRni90Ofkx1KTHNu/keAMrZn+u20Co30CpY6ti9pVPuTNPTwldhWVtfgcDp+XvM4Cj8/d24FLHG9gx+3PdBkL9BkodWxWzr3xKvXoqIv4VEf+OZfcZTGLZ82g+CWyu9IL0S0k3uPXqLWL9PV5/j9mf69bsWK2KORDq2KqYfUYzMhP5KZ6kR3t3PHl2HOluyfdQ8XRXx3PM/ly3gVC/gVLHVsVs9adZ92ksIb3p6gVgm5yBTwKWRMRtkZ/u6niO2eQ4rYrZ3+vXingDKWZrNSs7kW5cWUJ6d8SnHc8x+0KcVsXs7/UbKHVsVcxWfpoXCN5MenrrKo7nmH0lTqti9vf6DZQ6tipmKz/KlTYzM+tWX3nKrZmZrQScNMzMrDAnDTMzK8xJw8zMCnPSMDOzwpw0zHpJ0mJJMyQ9IOleSV9Ueo1nV9OMlXRQs8po1ihOGma993pETIiILUnPINod+EY304wFnDRspeP7NMx6SdKrEbFmRfdbgKnASGBj4GJgjTz48xFxh6S7gM2Bx4ELgR8BpwM7k56e+pOIOKdplTAryEnDrJeqk0buNx94G/AK6TlEb0gaT3pMdruknYEvR8QeefwjSC/lOUXSKsDtwP4R8XhTK2PWjSGtLoBZP9XxGOyhwFmSJgCLSe+RruWDpAfefTR3DwfGk45EzPoMJw2zBsvNU4uBOaRzG88DbyedQ3yjs8lIL+65vimFNOshnwg3ayBJo4CfAmdFavsdDjwXEUuAQ0jvX4DUbDWsYtLrgc9JGprns5mkNTDrY3ykYdZ7q0maQWqKWkQ68f39POxs4DeS9gf+BLyW+98HLJJ0L3AB8EPSFVX35De8zQX2aVYFzIryiXAzMyvMzVNmZlaYk4aZmRXmpGFmZoU5aZiZWWFOGmZmVpiThpmZFeakYWZmhf1/j5GgS9l2LLYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(216,)\n",
      "(216,)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(451,)\n",
      "(451,)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "repos = combine.groupby('repo_link')\n",
    "x = 0\n",
    "for repo in repos:\n",
    "    fin_dates = []\n",
    "    fin_values = []\n",
    "    dates = []\n",
    "    values = []\n",
    "    mail_list = None\n",
    "    j = 0\n",
    "    rep = (combine.loc[combine['repo_link'] == repo[0]])\n",
    "    grouped = rep.groupby('subject').groups\n",
    "    for group in grouped:\n",
    "        #print(group)\n",
    "        dates = (combine.loc[combine['subject'] == group]['date']).tolist()\n",
    "        values = []\n",
    "        for val in dates:\n",
    "            #print(val)\n",
    "            values.append((combine.loc[combine['date'] == val]['score']).tolist()[0])\n",
    "        #print(\"Issue number: \",issue_num[0])\n",
    "        x = np.array(dates)\n",
    "        x = [str(k) for k in x]\n",
    "        dates = [dt.datetime.strptime(d[:10],'%Y-%m-%d').date() for d in x]\n",
    "        dates1 = [x for x in dates]\n",
    "        values = sum(values[0:len(values)])/len(values)\n",
    "        if(values >= -0.5):\n",
    "            values = 0\n",
    "        else:\n",
    "            values = -1\n",
    "        fin_dates.append(dates1[0])\n",
    "        fin_values.append(values)\n",
    "        num+=1\n",
    "        j+=1\n",
    "    fin_dates = mdates.date2num(fin_dates)\n",
    "    fin_values = np.asarray(fin_values)\n",
    "    print(fin_dates.shape)\n",
    "    print(fin_values.shape)\n",
    "    plot.plot_date(fin_dates, fin_values)\n",
    "    #plot.plot(fin_dates,fin_values)\n",
    "    plot.gcf().autofmt_xdate()\n",
    "    plot.title(\"Pull Requests: \"+ str(repo[0]))\n",
    "    plot.ylabel('Score')\n",
    "    plot.xlabel('Date')\n",
    "    #print(fin_dates)\n",
    "    slope, intercept, r_value, p_value, std_err = stats.linregress(fin_dates,fin_values)\n",
    "    line = slope*fin_dates+intercept\n",
    "    plot.plot(fin_dates,fin_values,'o', fin_dates, line)\n",
    "    #plot.savefig('foo' + str(x) + '.png')\n",
    "    #x+=1\n",
    "    plot.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (augur)",
   "language": "python",
   "name": "augur"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}