Model7.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import findspark\n",
    "findspark.init('/home/sdmohant/spark-3.0.0-bin-hadoop2.7')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pyspark\n",
    "from pyspark import SparkContext,SparkConf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.sql import SparkSession"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sdmohant/spark-3.0.0-bin-hadoop2.7/python/pyspark/context.py:220: DeprecationWarning: Support for Python 2 and Python 3 prior to version 3.6 is deprecated as of Spark 3.0. See also the plan for dropping Python 2 support at https://spark.apache.org/news/plan-for-dropping-python-2-support.html.\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "#spark = SparkSession.builder.appName(\"MyApp\").getOrCreate()\n",
    "#spark = SparkSession.builder.appName(\"MyName\").getOrCreate()\n",
    "conf = SparkConf().set(\"spark.cores.max\", \"16\") \\\n",
    "    .set(\"spark.driver.memory\", \"16g\") \\\n",
    "    .set(\"spark.executor.memory\", \"16g\") \\\n",
    "    .set(\"spark.executor.memory_overhead\", \"16g\") \\\n",
    "    .set(\"spark.driver.maxResultsSize\", \"0\") \\\n",
    "    .set(\"spark.sql.shuffle.partitions\",\"300\")\n",
    "\n",
    "sc = SparkContext(appName=\"model7\", conf=conf)\n",
    "spark = SparkSession(sc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.sql.functions import isnan, count, col, when"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df1 = spark.read.options(header = True, inferSchema = True,delimiter=',').csv(\"AutPapJouCSCI.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = spark.read.options(header = True, inferSchema = True,delimiter=',').csv(\"DataAfterModel-6.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df1.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "root\n",
      " |-- PaperID: long (nullable = true)\n",
      " |-- PaperRefCnt: integer (nullable = true)\n",
      " |-- avg(PaperID): double (nullable = true)\n",
      " |-- avg(AffiliationRank): double (nullable = true)\n",
      " |-- avg(AfPaperCount): double (nullable = true)\n",
      " |-- avg(AfCitationCount): double (nullable = true)\n",
      " |-- Paper_Rank: integer (nullable = true)\n",
      " |-- Paper_RefCnt: integer (nullable = true)\n",
      " |-- CitationCount: integer (nullable = true)\n",
      " |-- EstimatedCount: integer (nullable = true)\n",
      " |-- Journal_Rank: integer (nullable = true)\n",
      " |-- Journal_Pap_Cnt: integer (nullable = true)\n",
      " |-- Journal_Cit_Cnt: integer (nullable = true)\n",
      " |-- CSRank: integer (nullable = true)\n",
      " |-- CSPaperCount: integer (nullable = true)\n",
      " |-- CSCitation: integer (nullable = true)\n",
      " |-- CIRank: integer (nullable = true)\n",
      " |-- CIPaperCount: integer (nullable = true)\n",
      " |-- CIPaperCitationCnt: integer (nullable = true)\n",
      " |-- DocTypeIndex: double (nullable = true)\n",
      " |-- max_AuthorRank: integer (nullable = true)\n",
      " |-- min_AutPaperCount: integer (nullable = true)\n",
      " |-- max_AutPaperCount: integer (nullable = true)\n",
      " |-- min_AutCitationCount: integer (nullable = true)\n",
      " |-- max_AutCitationCount: integer (nullable = true)\n",
      " |-- min_AuthorRank: integer (nullable = true)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "df2.printSchema()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import matplotlib.pyplot as plt \n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = df2.toPandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "corr = df2.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "26"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df2.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#corr = corr.style.background_gradient(cmap='coolwarm', axis=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2.drop(['PaperID', 'EstimatedCount','avg(PaperID)'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = df2[[\"PaperRefCnt\", \"avg(AffiliationRank)\",\"avg(AfPaperCount)\",\"avg(AfCitationCount)\",\"Paper_Rank\",\"Paper_RefCnt\",\"Journal_Rank\", \\\n",
    "          \"Journal_Pap_Cnt\",\"Journal_Cit_Cnt\",\"CSRank\",\"CSPaperCount\",\"CSCitation\",\"CIRank\",\"CIPaperCount\",\"CIPaperCitationCnt\",\\\n",
    "          \"DocTypeIndex\",\"max_AuthorRank\",\"min_AutPaperCount\",\"max_AutPaperCount\",\"min_AutCitationCount\",\"max_AutCitationCount\",\\\n",
    "          \"min_AuthorRank\",\"CitationCount\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(76900, 23)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "list"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bins=df2['CitationCount'].quantile([0,.05,0.1,0.15,0.20,0.25,0.3,0.35,0.40,0.45,0.5,0.55,0.6,0.65,0.70,0.75,0.80,0.85,0.90,0.95,1]).tolist()\n",
    "type(bins)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/sdmohant/.virtualenvs/datascience/local/lib/python2.7/site-packages/matplotlib/axes/_axes.py:6571: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "/home/sdmohant/.virtualenvs/datascience/local/lib/python2.7/site-packages/numpy/lib/histograms.py:823: RuntimeWarning: invalid value encountered in true_divide\n",
      "  return n/db/n.sum(), bin_edges\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(array([           nan,            nan,            nan,            nan,\n",
       "        2.00325098e-01,            nan, 1.12509753e-01, 8.62158648e-02,\n",
       "                   nan, 6.62418726e-02, 5.35630689e-02, 4.44603381e-02,\n",
       "        3.41092328e-02, 2.52925878e-02, 1.83008236e-02, 1.22782835e-02,\n",
       "        7.76549632e-03, 4.49501517e-03, 1.86485575e-03, 1.16690934e-05]),\n",
       " array([0.000e+00, 0.000e+00, 0.000e+00, 0.000e+00, 0.000e+00, 1.000e+00,\n",
       "        1.000e+00, 2.000e+00, 3.000e+00, 3.000e+00, 4.000e+00, 5.000e+00,\n",
       "        6.000e+00, 8.000e+00, 1.000e+01, 1.300e+01, 1.800e+01, 2.400e+01,\n",
       "        3.600e+01, 6.300e+01, 4.408e+03]),\n",
       " <a list of 1 Patch objects>)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(df2['CitationCount'], bins=bins, normed=True, alpha=0.5, histtype='stepfilled', color='steelblue', edgecolor='none')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7fd2d185da10>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df2.hist(column=\"CitationCount\", bins=[0,20,40,60,80,100,250,500,1000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = df2.iloc[0:60000, 0:22].values\n",
    "y_train = df2.iloc[:60000, 22].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = df2.iloc[60000:, 0:22].values\n",
    "y_test = df2.iloc[60000:, 22].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature Scaling\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
       "           max_features='auto', max_leaf_nodes=None,\n",
       "           min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "           min_samples_leaf=1, min_samples_split=2,\n",
       "           min_weight_fraction_leaf=0.0, n_estimators=20, n_jobs=None,\n",
       "           oob_score=False, random_state=0, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "regressor = RandomForestRegressor(n_estimators=20, random_state=0)\n",
    "regressor.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = regressor.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('Mean Absolute Error:', 4.596316568047337)\n",
      "('Mean Squared Error:', 860.8895785502958)\n",
      "('Root Mean Squared Error:', 29.340919865442117)\n"
     ]
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "\n",
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))\n",
    "\n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))\n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df2.iloc[:, 0:22].values\n",
    "y = df2.iloc[:, 22].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "#heights, bin_bounds = np.histogram(y_train, bins=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "plt.rcParams.update({'figure.figsize':(7,5), 'figure.dpi':100})\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_test, bins = [0,20,40,60,80,100,250,500,1000]) \n",
    "plt.title(\"Citation Count in Test Set\") \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_train, bins = [0,20,40,60,80,100,250,500,1000]) \n",
    "plt.title(\"Citation count in Train Set\") \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "#x = np.random.normal(size = 1000)\n",
    "#kwargs = dict(alpha=0.5, bins=100, density=True, stacked=True)\n",
    "#plt.hist(x1, **kwargs, color='g', label='Ideal')\n",
    "#plt.hist(y_train,*kwargs, color='g', label='Ideal')\n",
    "#plt.gca().set(title='Frequency Histogram of Citations', ylabel='Frequency')\n",
    "#plt.xlim(50,75)\n",
    "#plt.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Feature Scaling\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "regressor = RandomForestRegressor(n_estimators=20, random_state=0)\n",
    "regressor.fit(X_train, y_train)\n",
    "y_pred = regressor.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('Mean Absolute Error:', 4.106512787169484)\n",
      "('Mean Squared Error:', 408.5626383831816)\n",
      "('Root Mean Squared Error:', 20.212932453832167)\n"
     ]
    }
   ],
   "source": [
    "print('Mean Absolute Error:', metrics.mean_absolute_error(y_test, y_pred))\n",
    "print('Mean Squared Error:', metrics.mean_squared_error(y_test, y_pred))\n",
    "print('Root Mean Squared Error:', np.sqrt(metrics.mean_squared_error(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature: 0, Score: 0.00527\n",
      "Feature: 1, Score: 0.00398\n",
      "Feature: 2, Score: 0.00321\n",
      "Feature: 3, Score: 0.00485\n",
      "Feature: 4, Score: 0.85748\n",
      "Feature: 5, Score: 0.00695\n",
      "Feature: 6, Score: 0.00463\n",
      "Feature: 7, Score: 0.00837\n",
      "Feature: 8, Score: 0.00332\n",
      "Feature: 9, Score: 0.00354\n",
      "Feature: 10, Score: 0.00262\n",
      "Feature: 11, Score: 0.00285\n",
      "Feature: 12, Score: 0.00000\n",
      "Feature: 13, Score: 0.00549\n",
      "Feature: 14, Score: 0.00527\n",
      "Feature: 15, Score: 0.00029\n",
      "Feature: 16, Score: 0.04787\n",
      "Feature: 17, Score: 0.00742\n",
      "Feature: 18, Score: 0.00385\n",
      "Feature: 19, Score: 0.01532\n",
      "Feature: 20, Score: 0.00414\n",
      "Feature: 21, Score: 0.00328\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot\n",
    "# get importance\n",
    "importance = regressor.feature_importances_\n",
    "# summarize feature importance\n",
    "for i,v in enumerate(importance):\n",
    "\tprint('Feature: %0d, Score: %.5f' % (i,v))\n",
    "# plot feature importance\n",
    "pyplot.bar([x for x in range(len(importance))], importance)\n",
    "pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PaperRefCnt</th>\n",
       "      <th>avg(AffiliationRank)</th>\n",
       "      <th>avg(AfPaperCount)</th>\n",
       "      <th>avg(AfCitationCount)</th>\n",
       "      <th>Paper_Rank</th>\n",
       "      <th>Paper_RefCnt</th>\n",
       "      <th>Journal_Rank</th>\n",
       "      <th>Journal_Pap_Cnt</th>\n",
       "      <th>Journal_Cit_Cnt</th>\n",
       "      <th>CSRank</th>\n",
       "      <th>...</th>\n",
       "      <th>CIPaperCount</th>\n",
       "      <th>CIPaperCitationCnt</th>\n",
       "      <th>DocTypeIndex</th>\n",
       "      <th>max_AuthorRank</th>\n",
       "      <th>min_AutPaperCount</th>\n",
       "      <th>max_AutPaperCount</th>\n",
       "      <th>min_AutCitationCount</th>\n",
       "      <th>max_AutCitationCount</th>\n",
       "      <th>min_AuthorRank</th>\n",
       "      <th>CitationCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>20</td>\n",
       "      <td>7573.5</td>\n",
       "      <td>144863.0</td>\n",
       "      <td>6181726.75</td>\n",
       "      <td>19535</td>\n",
       "      <td>20</td>\n",
       "      <td>8166</td>\n",
       "      <td>4503</td>\n",
       "      <td>203353</td>\n",
       "      <td>10367</td>\n",
       "      <td>...</td>\n",
       "      <td>457</td>\n",
       "      <td>7357</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18422</td>\n",
       "      <td>6</td>\n",
       "      <td>342</td>\n",
       "      <td>150</td>\n",
       "      <td>7999</td>\n",
       "      <td>14057</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55</td>\n",
       "      <td>6247.0</td>\n",
       "      <td>288528.0</td>\n",
       "      <td>20686665.00</td>\n",
       "      <td>17891</td>\n",
       "      <td>55</td>\n",
       "      <td>9072</td>\n",
       "      <td>9299</td>\n",
       "      <td>121450</td>\n",
       "      <td>7775</td>\n",
       "      <td>...</td>\n",
       "      <td>702</td>\n",
       "      <td>16534</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16606</td>\n",
       "      <td>8</td>\n",
       "      <td>176</td>\n",
       "      <td>795</td>\n",
       "      <td>8024</td>\n",
       "      <td>14777</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   PaperRefCnt  avg(AffiliationRank)  avg(AfPaperCount)  avg(AfCitationCount)  \\\n",
       "0           20                7573.5           144863.0            6181726.75   \n",
       "1           55                6247.0           288528.0           20686665.00   \n",
       "\n",
       "   Paper_Rank  Paper_RefCnt  Journal_Rank  Journal_Pap_Cnt  Journal_Cit_Cnt  \\\n",
       "0       19535            20          8166             4503           203353   \n",
       "1       17891            55          9072             9299           121450   \n",
       "\n",
       "   CSRank      ...        CIPaperCount  CIPaperCitationCnt  DocTypeIndex  \\\n",
       "0   10367      ...                 457                7357           0.0   \n",
       "1    7775      ...                 702               16534           0.0   \n",
       "\n",
       "   max_AuthorRank  min_AutPaperCount  max_AutPaperCount  min_AutCitationCount  \\\n",
       "0           18422                  6                342                   150   \n",
       "1           16606                  8                176                   795   \n",
       "\n",
       "   max_AutCitationCount  min_AuthorRank  CitationCount  \n",
       "0                  7999           14057             19  \n",
       "1                  8024           14777            175  \n",
       "\n",
       "[2 rows x 23 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_new = spark.read.options(header = True, inferSchema = True,delimiter=',').csv(\"DataAfterModel-6.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Row(PaperID=9248908, PaperRefCnt=20, avg(PaperID)=9248908.0, avg(AffiliationRank)=7573.5, avg(AfPaperCount)=144863.0, avg(AfCitationCount)=6181726.75, Paper_Rank=19535, Paper_RefCnt=20, CitationCount=19, EstimatedCount=19, Journal_Rank=8166, Journal_Pap_Cnt=4503, Journal_Cit_Cnt=203353, CSRank=10367, CSPaperCount=6323, CSCitation=253240, CIRank=25000, CIPaperCount=457, CIPaperCitationCnt=7357, DocTypeIndex=0.0, max_AuthorRank=18422, min_AutPaperCount=6, max_AutPaperCount=342, min_AutCitationCount=150, max_AutCitationCount=7999, min_AuthorRank=14057)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['PaperID',\n",
       " 'PaperRefCnt',\n",
       " 'avg(PaperID)',\n",
       " 'avg(AffiliationRank)',\n",
       " 'avg(AfPaperCount)',\n",
       " 'avg(AfCitationCount)',\n",
       " 'Paper_Rank',\n",
       " 'Paper_RefCnt',\n",
       " 'CitationCount',\n",
       " 'EstimatedCount',\n",
       " 'Journal_Rank',\n",
       " 'Journal_Pap_Cnt',\n",
       " 'Journal_Cit_Cnt',\n",
       " 'CSRank',\n",
       " 'CSPaperCount',\n",
       " 'CSCitation',\n",
       " 'CIRank',\n",
       " 'CIPaperCount',\n",
       " 'CIPaperCitationCnt',\n",
       " 'DocTypeIndex',\n",
       " 'max_AuthorRank',\n",
       " 'min_AutPaperCount',\n",
       " 'max_AutPaperCount',\n",
       " 'min_AutCitationCount',\n",
       " 'max_AutCitationCount',\n",
       " 'min_AuthorRank']"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final = df_new.withColumnRenamed(\"CitationCount\",\"label\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "drop_cols = [\"PaperID\",\"avg(PaperID)\",\"EstimatedCount\"]\n",
    "df_final = df_final.drop(*drop_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.feature import VectorAssembler, VectorIndexer\n",
    "feature_list = []\n",
    "for col in df_final.columns:\n",
    "    if col == 'label':\n",
    "        continue\n",
    "    else:\n",
    "        feature_list.append(col)\n",
    "# Concatenates all feature columns into a single feature vector in a new column \"rawFeatures\"\n",
    "vectorAssembler = VectorAssembler(inputCols=feature_list, outputCol=\"features\")\n",
    "# Identifies categorical features and indexes them\n",
    "#vectorIndexer = VectorIndexer(inputCol=\"rawFeatures\", outputCol=\"features\", maxCategories=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.tuning import CrossValidator, ParamGridBuilder\n",
    "from pyspark.ml.evaluation import RegressionEvaluator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.regression import RandomForestRegressor\n",
    "\n",
    "rf = RandomForestRegressor(labelCol=\"label\", featuresCol=\"features\")\n",
    "#rf = RandomForestRegressor(featuresCol=\"features\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml import Pipeline\n",
    "pipeline = Pipeline(stages=[vectorAssembler, rf])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyspark.ml.tuning import ParamGridBuilder\n",
    "import numpy as np\n",
    "\n",
    "paramGrid = ParamGridBuilder() \\\n",
    "    .addGrid(rf.numTrees, [int(x) for x in np.linspace(start = 10, stop = 50, num = 3)]) \\\n",
    "    .addGrid(rf.maxDepth, [int(x) for x in np.linspace(start = 5, stop = 25, num = 3)]) \\\n",
    "    .build()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "crossval = CrossValidator(estimator=pipeline,\n",
    "                          estimatorParamMaps=paramGrid,\n",
    "                          evaluator=RegressionEvaluator(),\n",
    "                          numFolds=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "(train, test) = df_final.randomSplit([0.8, 0.2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "cvModel = crossval.fit(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Feature Importances')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 700x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "bestPipeline = cvModel.bestModel\n",
    "bestModel = bestPipeline.stages[1]\n",
    "\n",
    "importances = bestModel.featureImportances\n",
    "\n",
    "x_values = list(range(len(importances)))\n",
    "\n",
    "plt.bar(x_values, importances, orientation = 'vertical')\n",
    "plt.xticks(x_values, feature_list, rotation=80)\n",
    "plt.ylabel('Importance')\n",
    "plt.xlabel('Feature')\n",
    "plt.title('Feature Importances')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.15rc1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}