datascrum-sales-altdata-modelling-sample-v2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicting Company Sales vs with Alternative Data (web traffic)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import psycopg2\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "import datetime as dt\n",
    "import statsmodels.api as sm\n",
    "from sklearn import preprocessing\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Import the data from Postgresql as dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'user': 'altcap_usr', 'dbname': 'altcap_db', 'host': '167.99.88.131', 'port': '5432', 'tty': '', 'options': '', 'sslmode': 'prefer', 'sslcompression': '0', 'krbsrvname': 'postgres', 'target_session_attrs': 'any'} \n",
      "\n",
      "PostgreSQL connection is closed\n"
     ]
    }
   ],
   "source": [
    "DB_NAME = ''\n",
    "DB_USER = ''\n",
    "DB_PWD = ''\n",
    "DB_HOST = ''\n",
    "DB_PORT = ''\n",
    "\n",
    "try:\n",
    "    connection = psycopg2.connect(user = DB_USER,\n",
    "                                  password = DB_PWD,\n",
    "                                  host = DB_PWD ,\n",
    "                                  port = DB_PORT,\n",
    "                                  database = DB_NAME)\n",
    "    cursor = connection.cursor()\n",
    "    print ( connection.get_dsn_parameters(),\"\\n\")\n",
    "#     cursor.execute(\"SELECT company_id, date, traffic FROM trafficdata INNER JOIN salesestimate ON trafficdata.company_id=salesestimate.company_id;\")\n",
    "#     record = cursor.fetchall()\n",
    "#     record = cursor.execute(\"SELECT * FROM trafficdata;\")\n",
    "    record_traffic = pd.read_sql(\"SELECT * FROM trafficdata;\",connection)\n",
    "#     record_traffic = pd.DataFrame(cursor.fetchall())\n",
    "#     cursor.execute(\"SELECT * FROM eodfundamentalsdata;\")\n",
    "    record_eod = pd.read_sql(\"SELECT * FROM eodfundamentalsdata;\",connection)\n",
    "    record_sales = pd.read_sql(\"SELECT * FROM salesestimate;\",connection)\n",
    "#     print(\"You are connected to - \", record,\"\\n\")\n",
    "    \n",
    "except (Exception, psycopg2.Error) as error :\n",
    "    print (\"Error while connecting to PostgreSQL\", error)\n",
    "    \n",
    "finally:\n",
    "    #closing database connection.\n",
    "        if(connection):\n",
    "            cursor.close()\n",
    "            connection.close()\n",
    "            print(\"PostgreSQL connection is closed\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "asos_traffic = record_traffic[record_traffic[\"company_id\"]==1]\n",
    "asos_sales = record_sales[record_sales[\"company_id\"]==1]\n",
    "# asos_sales.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/pandas/core/generic.py:5096: 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",
      "  self[name] = value\n"
     ]
    }
   ],
   "source": [
    "asos_sales.date = asos_sales.date.map(lambda x: x.strftime('%Y-%m-%d'))\n",
    "asos_sales_sum = asos_sales[[\"date\",\"analysis_mean\"]]\n",
    "# asos_sales = asos_sales.groupby([\"date\"])['analysis_mean'].sum()\n",
    "# asos_sales_sum.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "asos_traffic_daily = asos_traffic[asos_traffic[\"period\"]==\"daily\"]\n",
    "asos_traffic_daily.date = asos_traffic_daily[\"date\"].map(lambda x: x.strftime('%Y-%m-%d'))\n",
    "asos_traffic_sum = asos_traffic_daily[[\"date\",\"traffic\",\"period\"]]\n",
    "# asos_traffic_sum.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>date</th>\n",
       "      <th>analysis_mean</th>\n",
       "      <th>traffic</th>\n",
       "      <th>period</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2019-03-01</td>\n",
       "      <td>2.787650e+09</td>\n",
       "      <td>10687</td>\n",
       "      <td>daily</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2019-03-02</td>\n",
       "      <td>2.787650e+09</td>\n",
       "      <td>5785</td>\n",
       "      <td>daily</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2019-03-03</td>\n",
       "      <td>2.787650e+09</td>\n",
       "      <td>4674</td>\n",
       "      <td>daily</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2019-03-04</td>\n",
       "      <td>2.787650e+09</td>\n",
       "      <td>9087</td>\n",
       "      <td>daily</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2019-03-05</td>\n",
       "      <td>2.787650e+09</td>\n",
       "      <td>9092</td>\n",
       "      <td>daily</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         date  analysis_mean  traffic period\n",
       "0  2019-03-01   2.787650e+09    10687  daily\n",
       "1  2019-03-02   2.787650e+09     5785  daily\n",
       "2  2019-03-03   2.787650e+09     4674  daily\n",
       "3  2019-03-04   2.787650e+09     9087  daily\n",
       "4  2019-03-05   2.787650e+09     9092  daily"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_total = pd.merge(asos_sales_sum, asos_traffic_sum,on=\"date\",how='inner')\n",
    "df_total.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df_total[[\"traffic\"]].values.astype(float)\n",
    "min_max_scaler = preprocessing.MinMaxScaler()\n",
    "x_scaled = min_max_scaler.fit_transform(X)\n",
    "\n",
    "y = df_total[[\"analysis_mean\"]].values.astype(float)\n",
    "y_scaled = min_max_scaler.fit_transform(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.18111052])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = sm.OLS (x_scaled,y_scaled)\n",
    "results = model.fit()\n",
    "params = results.params\n",
    "params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 1)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(x_scaled,y_scaled)\n",
    "plt.xlabel(\"Daily Traffic\")\n",
    "plt.ylabel(\"Estimated Daily Sales\")\n",
    "plt.xlim(0,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>traffic</th>\n",
       "      <th>sales</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.093448</td>\n",
       "      <td>0.734955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.017266</td>\n",
       "      <td>0.734955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.734955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.068582</td>\n",
       "      <td>0.734955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.068660</td>\n",
       "      <td>0.734955</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    traffic     sales\n",
       "0  0.093448  0.734955\n",
       "1  0.017266  0.734955\n",
       "2  0.000000  0.734955\n",
       "3  0.068582  0.734955\n",
       "4  0.068660  0.734955"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_traffic_and_sales = pd.DataFrame({\"traffic\":x_scaled.flatten(),\"sales\":y_scaled.flatten()})\n",
    "df_traffic_and_sales.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>traffic</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sales</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.000000</th>\n",
       "      <td>0.019939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.139565</th>\n",
       "      <td>0.196422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.153649</th>\n",
       "      <td>0.070696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.158771</th>\n",
       "      <td>0.471942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.162612</th>\n",
       "      <td>0.126019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.224072</th>\n",
       "      <td>0.154734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.280410</th>\n",
       "      <td>0.022395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.300896</th>\n",
       "      <td>0.023296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.708067</th>\n",
       "      <td>0.112921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.734955</th>\n",
       "      <td>0.026681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.000000</th>\n",
       "      <td>0.066254</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           traffic\n",
       "sales             \n",
       "0.000000  0.019939\n",
       "0.139565  0.196422\n",
       "0.153649  0.070696\n",
       "0.158771  0.471942\n",
       "0.162612  0.126019\n",
       "0.224072  0.154734\n",
       "0.280410  0.022395\n",
       "0.300896  0.023296\n",
       "0.708067  0.112921\n",
       "0.734955  0.026681\n",
       "1.000000  0.066254"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_traffic_and_sales_sum = df_traffic_and_sales.groupby(\"sales\")[[\"traffic\"]].mean()\n",
    "type (df_traffic_and_sales_sum)\n",
    "df_traffic_and_sales_sum"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1c24e0d860>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(df_traffic_and_sales_sum.index, df_traffic_and_sales_sum.traffic)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1c24f1c048>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "z = np.abs(stats.zscore(df_traffic_and_sales_sum))\n",
    "df_traffic_and_sales_sum = df_traffic_and_sales_sum[(z<1.9).all(axis=1)]\n",
    "df_traffic_and_sales_sum\n",
    "plt.scatter(df_traffic_and_sales_sum.index,df_traffic_and_sales_sum)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generalised Additive Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticGAM                                                                                               \n",
      "=============================================== ==========================================================\n",
      "Distribution:                      BinomialDist Effective DoF:                                      0.8387\n",
      "Link Function:                        LogitLink Log Likelihood:                                       -inf\n",
      "Number of Samples:                           10 AIC:                                                   inf\n",
      "                                                AICc:                                                  inf\n",
      "                                                UBRE:                                               2.7172\n",
      "                                                Scale:                                                 1.0\n",
      "                                                Pseudo R-Squared:                                  -0.0068\n",
      "==========================================================================================================\n",
      "Feature Function                  Lambda               Rank         EDoF         P > x        Sig. Code   \n",
      "================================= ==================== ============ ============ ============ ============\n",
      "s(0)                              [0.6]                20                        9.20e-01                 \n",
      "intercept                                              1                         4.33e-01                 \n",
      "==========================================================================================================\n",
      "Significance codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n",
      "\n",
      "WARNING: Fitting splines and a linear function to a feature introduces a model identifiability problem\n",
      "         which can cause p-values to appear significant when they are not.\n",
      "\n",
      "WARNING: p-values calculated in this manner behave correctly for un-penalized models or models with\n",
      "         known smoothing parameters, but when smoothing parameters have been estimated, the p-values\n",
      "         are typically lower than they should be, meaning that the tests reject the null too readily.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/pygam/pygam.py:1112: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  r2['McFadden'] = full_ll/null_ll\n",
      "/anaconda3/lib/python3.7/site-packages/pygam/pygam.py:1113: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  r2['McFadden_adj'] = 1. - (full_ll - self.statistics_['edof'])/null_ll\n",
      "/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:3: UserWarning: KNOWN BUG: p-values computed in this summary are likely much smaller than they should be. \n",
      " \n",
      "Please do not make inferences based on these values! \n",
      "\n",
      "Collaborate on a solution, and stay up to date at: \n",
      "github.com/dswah/pyGAM/issues/163 \n",
      "\n",
      "  This is separate from the ipykernel package so we can avoid doing imports until\n"
     ]
    }
   ],
   "source": [
    "import pygam\n",
    "gam = pygam.LogisticGAM().fit(df_traffic_and_sales_sum['traffic'],df_traffic_and_sales_sum.index)\n",
    "gam.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduce EOD data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/pandas/core/generic.py:5096: 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",
      "  self[name] = value\n"
     ]
    },
    {
     "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>id</th>\n",
       "      <th>date</th>\n",
       "      <th>frequency</th>\n",
       "      <th>total_revenue</th>\n",
       "      <th>ebit</th>\n",
       "      <th>company_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>3968</td>\n",
       "      <td>2019-02-28</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>657250000.0</td>\n",
       "      <td>2250000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>3969</td>\n",
       "      <td>2018-11-30</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>657250000.0</td>\n",
       "      <td>2250000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>3970</td>\n",
       "      <td>2018-08-31</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>629600000.0</td>\n",
       "      <td>36100000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>3971</td>\n",
       "      <td>2018-05-31</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>629600000.0</td>\n",
       "      <td>36100000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>3972</td>\n",
       "      <td>2018-02-28</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>579050000.0</td>\n",
       "      <td>14850000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id        date  frequency  total_revenue        ebit  company_id\n",
       "19  3968  2019-02-28  quarterly    657250000.0   2250000.0           1\n",
       "20  3969  2018-11-30  quarterly    657250000.0   2250000.0           1\n",
       "21  3970  2018-08-31  quarterly    629600000.0  36100000.0           1\n",
       "22  3971  2018-05-31  quarterly    629600000.0  36100000.0           1\n",
       "23  3972  2018-02-28  quarterly    579050000.0  14850000.0           1"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "record_eod_quarterly = record_eod[record_eod.frequency=='quarterly']\n",
    "record_eod_quarterly.date = record_eod_quarterly.date.map(lambda x: x.strftime('%Y-%m-%d'))\n",
    "record_eod_quarterly.fillna(method='ffill')\n",
    "record_eod_quarterly.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>traffic</th>\n",
       "      <th>id</th>\n",
       "      <th>date</th>\n",
       "      <th>frequency</th>\n",
       "      <th>total_revenue</th>\n",
       "      <th>ebit</th>\n",
       "      <th>company_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>160817</td>\n",
       "      <td>3968</td>\n",
       "      <td>2019-02-28</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>657250000.0</td>\n",
       "      <td>2250000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>14226</td>\n",
       "      <td>3969</td>\n",
       "      <td>2018-11-30</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>657250000.0</td>\n",
       "      <td>2250000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>15588</td>\n",
       "      <td>3970</td>\n",
       "      <td>2018-08-31</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>629600000.0</td>\n",
       "      <td>36100000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>19720</td>\n",
       "      <td>3971</td>\n",
       "      <td>2018-05-31</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>629600000.0</td>\n",
       "      <td>36100000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>24221</td>\n",
       "      <td>3972</td>\n",
       "      <td>2018-02-28</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>579050000.0</td>\n",
       "      <td>14850000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    traffic    id        date  frequency  total_revenue        ebit  \\\n",
       "19   160817  3968  2019-02-28  quarterly    657250000.0   2250000.0   \n",
       "20    14226  3969  2018-11-30  quarterly    657250000.0   2250000.0   \n",
       "21    15588  3970  2018-08-31  quarterly    629600000.0  36100000.0   \n",
       "22    19720  3971  2018-05-31  quarterly    629600000.0  36100000.0   \n",
       "23    24221  3972  2018-02-28  quarterly    579050000.0  14850000.0   \n",
       "\n",
       "    company_id  \n",
       "19           1  \n",
       "20           1  \n",
       "21           1  \n",
       "22           1  \n",
       "23           1  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict_traffic={}\n",
    "for i,j in zip(record_eod_quarterly.loc[:,'date'], record_eod_quarterly.loc[20:,'date']):\n",
    "    total_traffic=asos_traffic[(asos_traffic['date'] < i) & (asos_traffic['date'] >=j)]['traffic'].sum()\n",
    "    dict_traffic.update({i:total_traffic})\n",
    "record_eod_summary=pd.DataFrame.from_dict(dict_traffic,orient='index')\n",
    "record_eod_summary.columns=['traffic']\n",
    "record=pd.merge(record_eod_summary,record_eod_quarterly,left_index=True,right_on='date',how='inner')\n",
    "record.fillna(method='ffill')\n",
    "record.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1c24790438>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAEDCAYAAAAVyO4LAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAFO5JREFUeJzt3X2QXXV9x/H3l2WBAGqgWS0EasTRtCBCcCsiraNIDSBqSnUGpu341MlYraOtjUOKUyjTGdS0HerUqqmiVhEExJSharQ+VJ2p0Q0hBMRoBNQs1CxjU58yNYRv/7jnJpvN7t5zb+7DOXvfr5md3P3ds/d+88vuJ2fP+T1EZiJJqo8jBl2AJKk9Brck1YzBLUk1Y3BLUs0Y3JJUMwa3JNVMz4I7Im6IiF0RcW+JY38jIr4cEVsi4p6IuKRXdUlS3fXyjPsjwEUlj30HcEtmrgAuB/65V0VJUt31LLgz86vAT6a3RcTTI+JzEbE5Ir4WEb/ZPBx4YvH4ScDDvapLkuruyD6/33rgDZn5vYg4l8aZ9QXANcDnI+LNwHHAhX2uS5Jqo2/BHRHHA88Hbo2IZvPRxZ9XAB/JzL+PiPOAj0XEszLz8X7VJ0l10c8z7iOA3Zl59izPvZ7ienhm/ldEHAMsAXb1sT5JqoW+DQfMzJ8CD0bEqwCi4azi6R8CLy7afws4BpjqV22SVCfRq9UBI+Im4IU0zpx/DFwNfAl4H3ASMArcnJnXRsTpwL8Ax9O4Ufn2zPx8TwqTpJrrWXBLknrDmZOSVDM9uTm5ZMmSXLZsWS9eWpIWpM2bNz+amWNlju1JcC9btoyJiYlevLQkLUgR8YOyx3qpRJJqxuCWpJoxuCWpZgxuSaoZg1uSaqbfqwNWwoYtk6zbuJ2Hd+/h5MWLWLNyOatWLB10WZJqqt+ZMnTBvWHLJGtv38aevfsAmNy9h7W3bwMwvCW1bRCZMnSXStZt3L6/g5v27N3Huo3bB1SRpDobRKYMXXA/vHtPW+2SNJ9BZMrQBffJixe11S5J8xlEpgxdcK9ZuZxFoyMHtS0aHWHNyuUDqkhSnQ0iU4bu5mTzZoGjSiR1wyAypSfrcY+Pj6eLTElSeRGxOTPHyxw7dJdKJKnuhu5SyXStBs07UUdSFQ1tcLcaNO9EHUlVNbSXSloNmneijqSqGtrgbjVo3ok6kqpqaIO71aB5J+pIqqqhDe5Wg+adqCOpqob25mSrQfNO1JFUVU7AkaQK6PoEnIhYHBG3RcR3IuL+iDjv8EqUJHWq7KWSfwQ+l5mvjIijgGN7WJMkaR4tgzsingi8AHgNQGb+CvhVb8uSJM2lzKWS04Ap4MMRsSUiPhgRx808KCJWR8RERExMTU11vVBJUkOZ4D4SOAd4X2auAH4BXDnzoMxcn5njmTk+NjbW5TIlSU1lgnsnsDMzNxWf30YjyCVJA9AyuDPzv4EfRURz5smLgW/3tCpJ0pzKjip5M3BjMaLkAeC1vSupc4NYhtWlXyX1W6ngzsy7gVIDwwdlEMuwuvSrpEFYMGuVDGIZVpd+lTQICya4B7EMq0u/ShqEBRPcg1iG1aVfJQ3CggnuQSzD6tKvkgZhwSzrOohlWF36VdIguKyrJFVA15d1lSRVh8EtSTVjcEtSzRjcklQzBrck1YzBLUk1Y3BLUs0Y3JJUMwa3JNWMwS1JNWNwS1LNGNySVDMGtyTVjMEtSTVT+fW4N2yZ5Jo77mP3nr0AnHDsKFe/7AzXvJY0tEoFd0Q8BPwM2Ac8VnbN2MO1Ycska27dyt7HD6wZ/j+/3Mua27YC7qQuaTi1c6nkRZl5dr9CGxo7y0wP7aa9+9Kd1CUNrUpf455vt3R3Upc0rMoGdwKfj4jNEbG6lwVNN99u6e6kLmlYlQ3u8zPzHOBi4E0R8YKZB0TE6oiYiIiJqamprhS3ZuVyRo+IQ9pHR8Kd1CUNrVLBnZkPF3/uAj4NPHeWY9Zn5nhmjo+NjXWluFUrlrLuVWexeNHo/rYTjh1l3SvP8sakpKHVclRJRBwHHJGZPysevwS4tueVFVatWGpIS9I0ZYYDPgX4dEQ0j/9EZn6up1VJkubUMrgz8wHgrD7UIkkqodLDASVJhzK4JalmDG5JqhmDW5JqxuCWpJqpxbKu6zZuZ3L3HkYi2JfJ0sWLWLNyueO7JQ2lSgf3hi2TrL19G3v27gNgXzZWCpzcvYe1t28DXNpV0vCp9KWSdRu37w/tmfbs3efSrpKGUqWDu9XSrS7tKmkYVTq4Wy3d6tKukoZRpYN7zcrlLBodmfW5RaMjLu0qaShV+uZk88ajo0ok6YBKBze4rKskzVTpSyWSpEMZ3JJUMwa3JNWMwS1JNWNwS1LNGNySVDMGtyTVjMEtSTVTOrgjYiQitkTEnb0sSJI0v3bOuN8C3N+rQiRJ5ZQK7og4BXgp8MHeliNJaqXsWiXXA28HntDDWtrS3NLs4d17ONlFpyQNkZZn3BFxKbArMze3OG51RExExMTU1FTXCpxNc0uzyd17SA5sZbZhy2RP31eSqqDMpZLzgZdHxEPAzcAFEfHxmQdl5vrMHM/M8bGxsS6XebDZtjRzKzNJw6JlcGfm2sw8JTOXAZcDX8rMP+p5ZfOYa8sytzKTNAxqOY57ri3L3MpM0jBoK7gz8yuZeWmviilrti3N3MpM0rCo/A44s5m+pZmjSiQNm1oGN7ilmaThVctr3JI0zAxuSaoZg1uSasbglqSaMbglqWYMbkmqGYNbkmrG4JakmjG4JalmDG5JqpnaTHkvu+ONO+NIWuhqEdzNHW+amyc0d7wBDgrlssdJUp3V4lJJ2R1v3BlH0jCoRXCX3fHGnXEkDYNaBHfZHW/cGUfSMKhFcJfd8cadcSQNg1rcnCy7440740gaBpGZXX/R8fHxnJiY6PrrStJCFRGbM3O8zLG1uFQiSTrA4Jakmml5jTsijgG+ChxdHH9bZl7d68Lqypmb9oHUa2VuTv4fcEFm/jwiRoGvR8RnM/MbPa6tdpy5aR9I/dDyUkk2/Lz4dLT46P4dzQXAmZv2gdQPpa5xR8RIRNwN7AK+kJmbZjlmdURMRMTE1NRUt+usBWdu2gdSP5QK7szcl5lnA6cAz42IZ81yzPrMHM/M8bGxsW7XWQvO3LQPpH5oa1RJZu4GvgJc1JNqas6Zm/aB1A9lRpWMAXszc3dELAIuBN7V88pqyJmb9oHUDy1nTkbEs4GPAiM0ztBvycxr5/uahTpzsl/D3BxOJw2fdmZOtjzjzsx7gBWHXVXN9WuYm8PpJLXizMmS+jXMzeF0kloxuEvq1zA3h9NJasXgLqlfw9wcTiepFYO7pH4Nc3M4naRWarGRQhX0a5ibw+kkteJGCpJUAW6kIEkL2NBcKun1pBYnzUjql6EI7l5PanHSjKR+WvCXSjZsmeRtt2yddVLL227ZytOu/HfOf+eX2LBlsuP3cNKMpH5a0GfczTPhfXPcgG22H+4ZspNmJPXTgj7jnu1MeC6Hc4bspBlJ/bSgg7vdM95Oz5CdNCOpnxZ0cLd7xtvpGfKqFUu57rIzWbp4EQEsXbyI6y470xuTknpiQV/jXrNy+UGjPaBxJvwHz1nKpzZPHtJ+OGfIq1YsNagl9cWCDu75po+PP/VEx11LqiWnvEtSBTjlXZIWMINbkmqm9te4q7BGSBVqkDQ8ah3cVVgjpAo1SBouLS+VRMSpEfHliLg/Iu6LiLf0o7AyqrBGSBVqkDRcypxxPwa8LTPviognAJsj4guZ+e0e19ZSFdYIqUINkoZLyzPuzHwkM+8qHv8MuB+oxDWAKqwRUoUaJA2XtkaVRMQyYAWwaZbnVkfERERMTE1Ndae6FqqwRkgVapA0XErfnIyI44FPAW/NzJ/OfD4z1wProTEBp2sVzqMKG+tWoQZJw6XUzMmIGAXuBDZm5j+0Ot6Zk5LUnnZmTrY8446IAD4E3F8mtBe6bo7Zdvy3pE6UuVRyPvDHwLaIuLto+6vM/Ezvyqqmbo7Zdvy3pE6VGVXy9cyMzHx2Zp5dfAxdaEN3x2w7/ltSp1yrpA3dHLPt+G9JnTK429DNMduO/5bUKYO7Dd0cs+34b0mdqvUiU/3WzTHbjv+W1Cl3wJGkCnAHHElawAxuSaoZg1uSasbglqSaqcyokm6s2+HaH5KGQSWCe8OWSdbcupW9jzdGuEzu3sOaW7cC5dft6MXaH/5HIKmKKnGp5Jo77tsf2k17H0+uueO+0q/R7bU/mv8RTO7eQ3LgP4INWyY7ej1J6pZKBPfuPXvbap9Nt9f+cBEoSVVVieDuhm6v/eEiUJKqqhLBfcKxo221z6bba3+4CJSkqqpEcF/9sjMYHYmD2kZHgqtfdkbp11i1YinXXXYmSxcvIoClixdx3WVndnwz0UWgJFVVJUaVdGvBpVUrlnZt1IeLQEmqKheZkqQK6Opmwf3yjg3buGnTj9iXyUgEV5x7Kn+76sxBlyVJlVOJ4H7Hhm18/Bs/3P/5vsz9nxveknSwStycvGnTj9pql6Rh1jK4I+KGiNgVEff2qoh9c1xnn6tdkoZZmTPujwAX9bgOSVJJLYM7M78K/KSXRUSb7ZI0zLp2jTsiVkfERERMTE1NtfW1c10Q8UKJJB2qa8Gdmeszczwzx8fGxtr62pGY/dx6rnZJGmaVGFVyxbmnttUuScOsEuO4m2O1nYAjSa21nPIeETcBLwSWAD8Grs7MD833NU55l6T2dHXKe2ZecfglSZK6pRKXSsC1SiSprEoEt2uVSFJ5lRhV4lolklReJYLbtUokqbxKBLckqTyDW5JqphLB7ZR3SSqvEsF92tixbbVL0jCrRHDv2PWLttolaZhVIrhd1lWSyqtEcEuSyjO4JalmKhHcz3jycW21S9Iwq0RwPzD1y7baJWmYVSK4nfIuSeVVIrglSeVVIrjnmh/pvElJOlQlgttx3JJUXiWCW5JUnsEtSTVjcEtSzZQK7oi4KCK2R8SOiLiy10VJkubWMrgjYgR4L3AxcDpwRUSc3uvCJEmzK3PG/VxgR2Y+kJm/Am4GXtHbsiRJcykT3EuB6dut7yzaDhIRqyNiIiImpqamulWfJGmGMsE92zyYQ4ZYZ+b6zBzPzPGxsbHDr0ySNKsywb0TOHXa56cAD/emHElSK2WC+1vAMyLiaRFxFHA5cEc3i3jonS9tq12ShtmRrQ7IzMci4s+AjcAIcENm3tftQgxpSSqnZXADZOZngM/0uBZJUgnOnJSkmjG4JalmDG5JqhmDW5JqJrIH+zpGxBTwgw6/fAnwaBfL6ZYq1lXFmsC62mVd7VmodT01M0vNXuxJcB+OiJjIzPFB1zFTFeuqYk1gXe2yrvZYl5dKJKl2DG5JqpkqBvf6QRcwhyrWVcWawLraZV3tGfq6KneNW5I0vyqecUuS5mFwS1LdZGYlPoCLgO3ADuDKHr3HqcCXgfuB+4C3FO0nAl8Avlf8eULRHsB7ipruAc6Z9lqvLo7/HvDqae3PAbYVX/MeistRJWobAbYAdxafPw3YVLz+J4Gjivaji893FM8vm/Yaa4v27cDKw+1bYDFwG/Cdos/Oq0hf/Xnx73cvcBNwzCD6C7gB2AXcO62t5/0z13u0qGtd8e94D/BpYHGn/dBJX89V17Tn/pLGBi1LqtBfRfubi7//fcC7+91f8/4MtBt+vfigEVrfB04DjgK2Aqf34H1Oan4DAE8AvktjA+R3NzsauBJ4V/H4EuCzxTfR84BN074RHij+PKF43PwB/SaNgIviay8uWdtfAJ/gQHDfAlxePH4/8KfF4zcC7y8eXw58snh8etFvRxffKN8v+rXjvgU+CvxJ8fgoGkE+0L6isW3eg8Ciaf30mkH0F/AC4BwODsie989c79GirpcARxaP3zWtrrb7od2+nq+uov1UGstG/4ADwT3o/noR8B/A0cXnT+53f837c9DtcOzko+jsjdM+Xwus7cP7/hvwezT+lzypaDsJ2F48/gBwxbTjtxfPXwF8YFr7B4q2k4DvTGs/6Lh56jgF+CJwAXBn8Y33KAd+0Pb3T/ENfl7x+MjiuJjZZ83jOu1b4Ik0AjJmtA+6r5p7oJ5Y/P3vBFYOqr+AZRz8A9/z/pnrPeara8Zzvw/cONvfr1U/dPK92aouGr/VnQU8xIHgHmh/0QjbC2c5rq/9NddHVa5xl9qQuJsiYhmwgsavKE/JzEcAij+f3KKu+dp3ztLeyvXA24HHi89/DdidmY/N8jr737t4/n+L49uttZXTgCngwxGxJSI+GBHHMeC+ysxJ4O+AHwKP0Pj7b2bw/dXUj/6Z6z3Keh2NM9JO6urke3NOEfFyYDIzt854atD99UzgdyNiU0T8Z0T8dod1dbW/mqoS3KU2JO7am0UcD3wKeGtm/nS+Q2dpyw7a56vlUmBXZm4u8b59qalwJI1fH9+XmSuAX9D4NXMufakrIk4AXkHj19STgeOAi+d5rX71VyuVqCMirgIeA27sQV1t1RwRxwJXAX8929NdrKsTR9K4FPM8YA1wS0REl+vquOaqBHffNiSOiFEaoX1jZt5eNP84Ik4qnj+Jxo2K+eqar/2UWdrncz7w8oh4CLiZxuWS64HFEdHcoWj66+x/7+L5JwE/6aDWVnYCOzNzU/H5bTSCfJB9BXAh8GBmTmXmXuB24PkMvr+a+tE/c73HvCLi1cClwB9m8ft5B3U9Svt9PZen0/gPeGvx/X8KcFdE/HoHdXW7v3YCt2fDN2n8Nrykg7q62V8HlLme0usPGv+7PUDjH7F5Yf+MHrxPAP8KXD+jfR0H37x4d/H4pRx8g+SbRfuJNK7/nlB8PAicWDz3reLY5g2SS9qo74UcuDl5Kwff0Hhj8fhNHHxD45bi8RkcfNPkARo3TDruW+BrwPLi8TVFPw20r4BzadzlP7b4uo/SuPs/kP7i0GujPe+fud6jRV0XAd8GxmYc13Y/tNvX89U147mHOHCNe9D99Qbg2uLxM2lc0oh+99ec33fdCsXD/aBxF/m7NO7MXtWj9/gdGr+K3APcXXxcQuO60hdpDNf54rRvhADeW9S0DRif9lqvozGMZwfw2mnt4zSGqX0f+CdK3mwovvaFHAju02jcJd9R/MM3724fU3y+o3j+tGlff1XxvtuZNkKj074FzgYmiv7aUPygDLyvgL+hMbTtXuBjxQ9R3/uLxlDER4C9NM6eXt+P/pnrPVrUtYNG+DS/79/faT900tdz1TXj+Yc4eDjgIPvrKODjxevdBVzQ7/6a78Mp75JUM1W5xi1JKsnglqSaMbglqWYMbkmqGYNbkmrG4JakmjG4Jalm/h/A8n8P37oRugAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(record['traffic'],record['total_revenue'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have an outlier on the top right because this is calculated from the sum of daily values of traffic, whereas other points are calculated from monthly values. There might be problems with the daily values or unit difference. For simplicity, we discard this point for the time being. We also discard all values with zero because that means there is no data at that point."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>traffic</th>\n",
       "      <th>id</th>\n",
       "      <th>date</th>\n",
       "      <th>frequency</th>\n",
       "      <th>total_revenue</th>\n",
       "      <th>ebit</th>\n",
       "      <th>company_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>14226</td>\n",
       "      <td>3969</td>\n",
       "      <td>2018-11-30</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>657250000.0</td>\n",
       "      <td>2250000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>15588</td>\n",
       "      <td>3970</td>\n",
       "      <td>2018-08-31</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>629600000.0</td>\n",
       "      <td>36100000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>19720</td>\n",
       "      <td>3971</td>\n",
       "      <td>2018-05-31</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>629600000.0</td>\n",
       "      <td>36100000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>24221</td>\n",
       "      <td>3972</td>\n",
       "      <td>2018-02-28</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>579050000.0</td>\n",
       "      <td>14850000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>9077</td>\n",
       "      <td>3973</td>\n",
       "      <td>2017-11-30</td>\n",
       "      <td>quarterly</td>\n",
       "      <td>579050000.0</td>\n",
       "      <td>14850000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    traffic    id        date  frequency  total_revenue        ebit  \\\n",
       "20    14226  3969  2018-11-30  quarterly    657250000.0   2250000.0   \n",
       "21    15588  3970  2018-08-31  quarterly    629600000.0  36100000.0   \n",
       "22    19720  3971  2018-05-31  quarterly    629600000.0  36100000.0   \n",
       "23    24221  3972  2018-02-28  quarterly    579050000.0  14850000.0   \n",
       "24     9077  3973  2017-11-30  quarterly    579050000.0  14850000.0   \n",
       "\n",
       "    company_id  \n",
       "20           1  \n",
       "21           1  \n",
       "22           1  \n",
       "23           1  \n",
       "24           1  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "record_cleaned=record.loc[~(record==0).any(axis=1)]\n",
    "record_cleaned=record_cleaned.loc[20:,:]\n",
    "record_cleaned.fillna(method='ffill')\n",
    "record_cleaned.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Quarterly Revenue')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(record_cleaned['traffic'],record_cleaned['total_revenue'])\n",
    "plt.xlabel('Quarterly Traffic')\n",
    "plt.ylabel('Quarterly Revenue')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score, GridSearchCV\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def rfr_model(X, y):\n",
    "    gsc = GridSearchCV(\n",
    "        estimator=RandomForestRegressor(),\n",
    "        param_grid={\n",
    "            'max_depth': range(3,7),\n",
    "            'n_estimators': (10, 50, 100, 1000),\n",
    "        },\n",
    "        cv=5, scoring='neg_mean_squared_error', verbose=0, n_jobs=-1)\n",
    "    \n",
    "    grid_result = gsc.fit(X, y)\n",
    "    best_params = grid_result.best_params_\n",
    "    \n",
    "    rfr = RandomForestRegressor(max_depth=best_params[\"max_depth\"], n_estimators=best_params[\"n_estimators\"], random_state=False, verbose=False)\n",
    "\n",
    "    scores = cross_val_score(rfr, X, y, cv=10, scoring='neg_mean_absolute_error')\n",
    "\n",
    "    return scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_search.py:841: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_search.py:740: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  self.best_estimator_.fit(X, y, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n",
      "/anaconda3/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:528: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
      "  estimator.fit(X_train, y_train, **fit_params)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([-0.73715083, -0.49218351, -0.15415807, -0.12976833, -0.08906912,\n",
       "       -0.13422129, -0.65898628, -0.33086571, -0.21143793, -0.64996389])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "record_cleaned = record_cleaned.fillna(method='ffill')\n",
    "# X=np.array(record_cleaned.loc['traffic']).reshape(-1,1)\n",
    "X = record_cleaned[['traffic']].values.astype(float)\n",
    "X_scaled = min_max_scaler.fit_transform(X)\n",
    "# y=np.array(record_cleaned.total_revenue).reshape(-1,1)\n",
    "y = record_cleaned[['total_revenue']].values.astype(float)\n",
    "y_scaled = min_max_scaler.fit_transform(y)\n",
    "score=rfr_model(X_scaled,y_scaled)\n",
    "score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import pandas as pd\n",
    "import torch.nn as nn\n",
    "import torchvision.datasets as dsets\n",
    "import torchvision.transforms as transforms\n",
    "from torch.autograd import Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = torch.FloatTensor(X_scaled)\n",
    "y = torch.FloatTensor(y_scaled)\n",
    "x, y = Variable(x), Variable(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Net(nn.Module):\n",
    "    def __init__(self, input_size, hidden_size, num_classes):\n",
    "        super(Net, self).__init__()\n",
    "        self.fc1 = nn.Linear(input_size, hidden_size)\n",
    "        self.relu = nn.ReLU()\n",
    "        self.fc2 = nn.Linear(hidden_size, num_classes)\n",
    "        self.relu = nn.ReLU()\n",
    "    \n",
    "    def forward(self, x):\n",
    "        out = self.fc1 (x)\n",
    "        out = self.relu (out)\n",
    "        out = self.fc2 (out)\n",
    "        return out"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch[0], loss:0.374945\n",
      "Epoch[20], loss:0.335285\n",
      "Epoch[40], loss:0.298502\n",
      "Epoch[60], loss:0.264926\n",
      "Epoch[80], loss:0.234353\n",
      "Epoch[100], loss:0.206607\n",
      "Epoch[120], loss:0.181598\n",
      "Epoch[140], loss:0.159218\n",
      "Epoch[160], loss:0.139333\n",
      "Epoch[180], loss:0.122021\n",
      "Epoch[200], loss:0.107176\n",
      "Epoch[220], loss:0.094492\n",
      "Epoch[240], loss:0.083788\n",
      "Epoch[260], loss:0.074909\n",
      "Epoch[280], loss:0.067649\n",
      "Epoch[300], loss:0.061788\n",
      "Epoch[320], loss:0.057122\n",
      "Epoch[340], loss:0.053462\n",
      "Epoch[360], loss:0.050637\n",
      "Epoch[380], loss:0.048490\n",
      "Epoch[400], loss:0.046883\n",
      "Epoch[420], loss:0.045699\n",
      "Epoch[440], loss:0.044838\n",
      "Epoch[460], loss:0.044220\n",
      "Epoch[480], loss:0.043781\n",
      "Epoch[500], loss:0.043471\n",
      "Epoch[520], loss:0.043251\n",
      "Epoch[540], loss:0.043095\n",
      "Epoch[560], loss:0.042983\n",
      "Epoch[580], loss:0.042899\n",
      "Epoch[600], loss:0.042835\n",
      "Epoch[620], loss:0.042783\n",
      "Epoch[640], loss:0.042740\n",
      "Epoch[660], loss:0.042702\n",
      "Epoch[680], loss:0.042667\n",
      "Epoch[700], loss:0.042635\n",
      "Epoch[720], loss:0.042605\n",
      "Epoch[740], loss:0.042576\n",
      "Epoch[760], loss:0.042547\n",
      "Epoch[780], loss:0.042520\n",
      "Epoch[800], loss:0.042493\n",
      "Epoch[820], loss:0.042467\n",
      "Epoch[840], loss:0.042442\n",
      "Epoch[860], loss:0.042418\n",
      "Epoch[880], loss:0.042394\n",
      "Epoch[900], loss:0.042371\n",
      "Epoch[920], loss:0.042348\n",
      "Epoch[940], loss:0.042326\n",
      "Epoch[960], loss:0.042305\n",
      "Epoch[980], loss:0.042285\n",
      "Epoch[1000], loss:0.042265\n",
      "Epoch[1020], loss:0.042246\n",
      "Epoch[1040], loss:0.042227\n",
      "Epoch[1060], loss:0.042209\n",
      "Epoch[1080], loss:0.042192\n",
      "Epoch[1100], loss:0.042175\n",
      "Epoch[1120], loss:0.042159\n",
      "Epoch[1140], loss:0.042143\n",
      "Epoch[1160], loss:0.042129\n",
      "Epoch[1180], loss:0.042114\n",
      "Epoch[1200], loss:0.042100\n",
      "Epoch[1220], loss:0.042087\n",
      "Epoch[1240], loss:0.042074\n",
      "Epoch[1260], loss:0.042062\n",
      "Epoch[1280], loss:0.042050\n",
      "Epoch[1300], loss:0.042039\n",
      "Epoch[1320], loss:0.042028\n",
      "Epoch[1340], loss:0.042018\n",
      "Epoch[1360], loss:0.042008\n",
      "Epoch[1380], loss:0.041999\n",
      "Epoch[1400], loss:0.041990\n",
      "Epoch[1420], loss:0.041982\n",
      "Epoch[1440], loss:0.041974\n",
      "Epoch[1460], loss:0.041966\n",
      "Epoch[1480], loss:0.041959\n",
      "Epoch[1500], loss:0.041952\n",
      "Epoch[1520], loss:0.041946\n",
      "Epoch[1540], loss:0.041939\n",
      "Epoch[1560], loss:0.041933\n",
      "Epoch[1580], loss:0.041928\n",
      "Epoch[1600], loss:0.041923\n",
      "Epoch[1620], loss:0.041918\n",
      "Epoch[1640], loss:0.041913\n",
      "Epoch[1660], loss:0.041909\n",
      "Epoch[1680], loss:0.041904\n",
      "Epoch[1700], loss:0.041900\n",
      "Epoch[1720], loss:0.041897\n",
      "Epoch[1740], loss:0.041893\n",
      "Epoch[1760], loss:0.041890\n",
      "Epoch[1780], loss:0.041887\n",
      "Epoch[1800], loss:0.041884\n",
      "Epoch[1820], loss:0.041881\n",
      "Epoch[1840], loss:0.041878\n",
      "Epoch[1860], loss:0.041876\n",
      "Epoch[1880], loss:0.041873\n",
      "Epoch[1900], loss:0.041871\n",
      "Epoch[1920], loss:0.041869\n",
      "Epoch[1940], loss:0.041867\n",
      "Epoch[1960], loss:0.041864\n",
      "Epoch[1980], loss:0.041862\n",
      "Epoch[2000], loss:0.041860\n",
      "Epoch[2020], loss:0.041858\n",
      "Epoch[2040], loss:0.041856\n",
      "Epoch[2060], loss:0.041854\n",
      "Epoch[2080], loss:0.041852\n",
      "Epoch[2100], loss:0.041849\n",
      "Epoch[2120], loss:0.041847\n",
      "Epoch[2140], loss:0.041844\n",
      "Epoch[2160], loss:0.041842\n",
      "Epoch[2180], loss:0.041839\n",
      "Epoch[2200], loss:0.041836\n",
      "Epoch[2220], loss:0.041832\n",
      "Epoch[2240], loss:0.041828\n",
      "Epoch[2260], loss:0.041824\n",
      "Epoch[2280], loss:0.041819\n",
      "Epoch[2300], loss:0.041814\n",
      "Epoch[2320], loss:0.041808\n",
      "Epoch[2340], loss:0.041802\n",
      "Epoch[2360], loss:0.041795\n",
      "Epoch[2380], loss:0.041787\n",
      "Epoch[2400], loss:0.041779\n",
      "Epoch[2420], loss:0.041770\n",
      "Epoch[2440], loss:0.041760\n",
      "Epoch[2460], loss:0.041749\n",
      "Epoch[2480], loss:0.041737\n",
      "Epoch[2500], loss:0.041724\n",
      "Epoch[2520], loss:0.041710\n",
      "Epoch[2540], loss:0.041695\n",
      "Epoch[2560], loss:0.041680\n",
      "Epoch[2580], loss:0.041664\n",
      "Epoch[2600], loss:0.041647\n",
      "Epoch[2620], loss:0.041631\n",
      "Epoch[2640], loss:0.041615\n",
      "Epoch[2660], loss:0.041598\n",
      "Epoch[2680], loss:0.041565\n",
      "Epoch[2700], loss:0.041527\n",
      "Epoch[2720], loss:0.041490\n",
      "Epoch[2740], loss:0.041454\n",
      "Epoch[2760], loss:0.041419\n",
      "Epoch[2780], loss:0.041385\n",
      "Epoch[2800], loss:0.041352\n",
      "Epoch[2820], loss:0.041319\n",
      "Epoch[2840], loss:0.041288\n",
      "Epoch[2860], loss:0.041263\n",
      "Epoch[2880], loss:0.041246\n",
      "Epoch[2900], loss:0.041231\n",
      "Epoch[2920], loss:0.041217\n",
      "Epoch[2940], loss:0.041203\n",
      "Epoch[2960], loss:0.041190\n",
      "Epoch[2980], loss:0.041178\n",
      "Epoch[3000], loss:0.041165\n",
      "Epoch[3020], loss:0.041154\n",
      "Epoch[3040], loss:0.041143\n",
      "Epoch[3060], loss:0.041132\n",
      "Epoch[3080], loss:0.041122\n",
      "Epoch[3100], loss:0.041112\n",
      "Epoch[3120], loss:0.041102\n",
      "Epoch[3140], loss:0.041093\n",
      "Epoch[3160], loss:0.041085\n",
      "Epoch[3180], loss:0.041077\n",
      "Epoch[3200], loss:0.041069\n",
      "Epoch[3220], loss:0.041061\n",
      "Epoch[3240], loss:0.041054\n",
      "Epoch[3260], loss:0.041048\n",
      "Epoch[3280], loss:0.041041\n",
      "Epoch[3300], loss:0.041035\n",
      "Epoch[3320], loss:0.041029\n",
      "Epoch[3340], loss:0.041024\n",
      "Epoch[3360], loss:0.041019\n",
      "Epoch[3380], loss:0.041014\n",
      "Epoch[3400], loss:0.041009\n",
      "Epoch[3420], loss:0.041005\n",
      "Epoch[3440], loss:0.041001\n",
      "Epoch[3460], loss:0.040997\n",
      "Epoch[3480], loss:0.040993\n",
      "Epoch[3500], loss:0.040989\n",
      "Epoch[3520], loss:0.040986\n",
      "Epoch[3540], loss:0.040983\n",
      "Epoch[3560], loss:0.040980\n",
      "Epoch[3580], loss:0.040977\n",
      "Epoch[3600], loss:0.040975\n",
      "Epoch[3620], loss:0.040972\n",
      "Epoch[3640], loss:0.040970\n",
      "Epoch[3660], loss:0.040968\n",
      "Epoch[3680], loss:0.040966\n",
      "Epoch[3700], loss:0.040964\n",
      "Epoch[3720], loss:0.040963\n",
      "Epoch[3740], loss:0.040961\n",
      "Epoch[3760], loss:0.040959\n",
      "Epoch[3780], loss:0.040958\n",
      "Epoch[3800], loss:0.040956\n",
      "Epoch[3820], loss:0.040955\n",
      "Epoch[3840], loss:0.040954\n",
      "Epoch[3860], loss:0.040953\n",
      "Epoch[3880], loss:0.040951\n",
      "Epoch[3900], loss:0.040950\n",
      "Epoch[3920], loss:0.040949\n",
      "Epoch[3940], loss:0.040948\n",
      "Epoch[3960], loss:0.040947\n",
      "Epoch[3980], loss:0.040946\n",
      "Epoch[4000], loss:0.040946\n",
      "Epoch[4020], loss:0.040945\n",
      "Epoch[4040], loss:0.040944\n",
      "Epoch[4060], loss:0.040943\n",
      "Epoch[4080], loss:0.040943\n",
      "Epoch[4100], loss:0.040942\n",
      "Epoch[4120], loss:0.040941\n",
      "Epoch[4140], loss:0.040941\n",
      "Epoch[4160], loss:0.040940\n",
      "Epoch[4180], loss:0.040939\n",
      "Epoch[4200], loss:0.040939\n",
      "Epoch[4220], loss:0.040938\n",
      "Epoch[4240], loss:0.040937\n",
      "Epoch[4260], loss:0.040937\n",
      "Epoch[4280], loss:0.040936\n",
      "Epoch[4300], loss:0.040936\n",
      "Epoch[4320], loss:0.040935\n",
      "Epoch[4340], loss:0.040934\n",
      "Epoch[4360], loss:0.040934\n",
      "Epoch[4380], loss:0.040933\n",
      "Epoch[4400], loss:0.040933\n",
      "Epoch[4420], loss:0.040932\n",
      "Epoch[4440], loss:0.040932\n",
      "Epoch[4460], loss:0.040931\n",
      "Epoch[4480], loss:0.040930\n",
      "Epoch[4500], loss:0.040930\n",
      "Epoch[4520], loss:0.040929\n",
      "Epoch[4540], loss:0.040929\n",
      "Epoch[4560], loss:0.040928\n",
      "Epoch[4580], loss:0.040927\n",
      "Epoch[4600], loss:0.040927\n",
      "Epoch[4620], loss:0.040926\n",
      "Epoch[4640], loss:0.040926\n",
      "Epoch[4660], loss:0.040925\n",
      "Epoch[4680], loss:0.040924\n",
      "Epoch[4700], loss:0.040924\n",
      "Epoch[4720], loss:0.040923\n",
      "Epoch[4740], loss:0.040922\n",
      "Epoch[4760], loss:0.040922\n",
      "Epoch[4780], loss:0.040921\n",
      "Epoch[4800], loss:0.040920\n",
      "Epoch[4820], loss:0.040920\n",
      "Epoch[4840], loss:0.040919\n",
      "Epoch[4860], loss:0.040918\n",
      "Epoch[4880], loss:0.040918\n",
      "Epoch[4900], loss:0.040917\n",
      "Epoch[4920], loss:0.040916\n",
      "Epoch[4940], loss:0.040916\n",
      "Epoch[4960], loss:0.040915\n",
      "Epoch[4980], loss:0.040914\n"
     ]
    }
   ],
   "source": [
    "net = Net(input_size=1, hidden_size=15, num_classes=1)\n",
    "criterion = nn.SmoothL1Loss()\n",
    "optimizer = torch.optim.Adam(net.parameters(), lr=0.0005)\n",
    "\n",
    "for epoch in range (5000):\n",
    "    inputs = x\n",
    "    target = y\n",
    "    out = net (inputs)\n",
    "    loss = criterion (out, target)\n",
    "    optimizer.zero_grad()\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "  \n",
    "    if epoch%20 == 0:\n",
    "        print ('Epoch[{}], loss:{:.6f}'.format(epoch,loss.data))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(tensor([[0.4377],\n",
      "        [0.4230],\n",
      "        [0.3782],\n",
      "        [0.3295],\n",
      "        [0.4934],\n",
      "        [0.5540],\n",
      "        [0.5509],\n",
      "        [0.5438],\n",
      "        [0.5567],\n",
      "        [0.5349],\n",
      "        [0.5228],\n",
      "        [0.5442],\n",
      "        [0.4904],\n",
      "        [0.2249],\n",
      "        [0.3054],\n",
      "        [0.2325],\n",
      "        [0.3580],\n",
      "        [0.3148],\n",
      "        [0.4531],\n",
      "        [0.4295],\n",
      "        [0.4655],\n",
      "        [0.5079],\n",
      "        [0.3237],\n",
      "        [0.3494],\n",
      "        [0.2426],\n",
      "        [0.2044],\n",
      "        [0.4953],\n",
      "        [0.3814]], grad_fn=<AddmmBackward>), tensor([[1.0000],\n",
      "        [0.9471],\n",
      "        [0.9471],\n",
      "        [0.8503],\n",
      "        [0.8503],\n",
      "        [0.7105],\n",
      "        [0.7105],\n",
      "        [0.6142],\n",
      "        [0.6142],\n",
      "        [0.6142],\n",
      "        [0.4900],\n",
      "        [0.3804],\n",
      "        [0.3764],\n",
      "        [0.3163],\n",
      "        [0.3163],\n",
      "        [0.2686],\n",
      "        [0.2686],\n",
      "        [0.2143],\n",
      "        [0.2143],\n",
      "        [0.2028],\n",
      "        [0.2028],\n",
      "        [0.1338],\n",
      "        [0.1338],\n",
      "        [0.0860],\n",
      "        [0.0860],\n",
      "        [0.0860],\n",
      "        [0.0074],\n",
      "        [0.0000]])) tensor([[-0.5623],\n",
      "        [-0.5241],\n",
      "        [-0.5688],\n",
      "        [-0.5207],\n",
      "        [-0.3568],\n",
      "        [-0.1565],\n",
      "        [-0.1596],\n",
      "        [-0.0704],\n",
      "        [-0.0575],\n",
      "        [-0.0793],\n",
      "        [ 0.0328],\n",
      "        [ 0.1638],\n",
      "        [ 0.1140],\n",
      "        [-0.0914],\n",
      "        [-0.0109],\n",
      "        [-0.0360],\n",
      "        [ 0.0894],\n",
      "        [ 0.1006],\n",
      "        [ 0.2388],\n",
      "        [ 0.2268],\n",
      "        [ 0.2628],\n",
      "        [ 0.3741],\n",
      "        [ 0.1899],\n",
      "        [ 0.2634],\n",
      "        [ 0.1566],\n",
      "        [ 0.1184],\n",
      "        [ 0.4879],\n",
      "        [ 0.3814]], grad_fn=<SubBackward0>)\n"
     ]
    }
   ],
   "source": [
    "y_pred = net(x)\n",
    "print((y_pred,y),y_pred-y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}