{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "import logging\n",
    "import sagemaker\n",
    "import boto3\n",
    "from random import shuffle\n",
    "sns.set()\n",
    "# Better aesthetics - https://seaborn.pydata.org/tutorial/aesthetics.html\n",
    "sns.set_style(\"whitegrid\", {'axes.grid': False})\n",
    "sns.set_color_codes('dark')\n",
    "sns.set_context(\"paper\")\n",
    "sns.despine()\n",
    "%run ./utils.ipynb\n",
    "# %run ./google_maps.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "standardizing gender ....\n",
      "dummy coding gender ....\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/psycopg2/__init__.py:144: UserWarning: The psycopg2 wheel package will be renamed from release 2.8; in order to keep installing from binary please use \"pip install psycopg2-binary\" instead. For details see: <http://initd.org/psycopg/docs/install.html#binary-install-from-pypi>.\n",
      "  \"\"\")\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>fan_id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>male</th>\n",
       "      <th>female</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>01daeb57fd8dd128049c397ba2e527439b46130753f9d3...</td>\n",
       "      <td>female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>000b06a1be22ef0a8ff82f504ef3f81ff3522224234821...</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0c1bb04b557a632b1e57f6e76058e6435dabc8117cd6a0...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              fan_id  gender  age  male  \\\n",
       "0  01daeb57fd8dd128049c397ba2e527439b46130753f9d3...  female  NaN     0   \n",
       "1  000b06a1be22ef0a8ff82f504ef3f81ff3522224234821...    male  NaN     1   \n",
       "2  0c1bb04b557a632b1e57f6e76058e6435dabc8117cd6a0...    None  NaN     0   \n",
       "\n",
       "   female  \n",
       "0       1  \n",
       "1       0  \n",
       "2       0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "schema = 'a441ffb172866cb4928c84a73de403ca15da4a54cc535e704413ab621'\n",
    "bb=prepare_fan_demographics(schema)\n",
    "gr = bb.groupby(['gender']) .count()\n",
    "bb.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "getting list of unique fans ....\n",
      "getting max tickets for fan ....\n",
      "getting max transaction value for fan ....\n",
      "(16830, 3)\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>fan_id</th>\n",
       "      <th>max_tickets_per_event</th>\n",
       "      <th>max_transcaction_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cdb82cdec7a0cecbd818a8ea3a73ed4cfd71f39a9e82fc...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9086da8d98db97ad441689769e00d7f5f7a12ef0e8413a...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>970b995d3475ffa7e7dbf2fa62c93bfd0da3ae25444030...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>fab6cb66d2b260af8dce182d258c88e9ed28f5577586fe...</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c53b066ec8e623cdc7a51daced9ec8a5f2dbba8c1bfe6b...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>686715b70d286f896da78aae295369d14cf53758349985...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>f74adefa1e98a075ab00d81a20714ee895482231203f4b...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>43.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7c21ffb79a720ff3196d98eca63d0c83cbab4d8b016691...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>10d54173fdd5bf37db17203c278514901faf05ab612cd4...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3046e255a8e2c8c82a533b78e99b9e169269466d56baf3...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              fan_id  max_tickets_per_event  \\\n",
       "0  cdb82cdec7a0cecbd818a8ea3a73ed4cfd71f39a9e82fc...                    1.0   \n",
       "1  9086da8d98db97ad441689769e00d7f5f7a12ef0e8413a...                    NaN   \n",
       "2  970b995d3475ffa7e7dbf2fa62c93bfd0da3ae25444030...                    1.0   \n",
       "3  fab6cb66d2b260af8dce182d258c88e9ed28f5577586fe...                    7.0   \n",
       "4  c53b066ec8e623cdc7a51daced9ec8a5f2dbba8c1bfe6b...                    1.0   \n",
       "5  686715b70d286f896da78aae295369d14cf53758349985...                    1.0   \n",
       "6  f74adefa1e98a075ab00d81a20714ee895482231203f4b...                    1.0   \n",
       "7  7c21ffb79a720ff3196d98eca63d0c83cbab4d8b016691...                    1.0   \n",
       "8  10d54173fdd5bf37db17203c278514901faf05ab612cd4...                    1.0   \n",
       "9  3046e255a8e2c8c82a533b78e99b9e169269466d56baf3...                    NaN   \n",
       "\n",
       "   max_transcaction_value  \n",
       "0                     0.0  \n",
       "1                    18.0  \n",
       "2                     0.0  \n",
       "3                     0.0  \n",
       "4                     0.0  \n",
       "5                     0.0  \n",
       "6                    43.9  \n",
       "7                     0.0  \n",
       "8                     0.0  \n",
       "9                     5.0  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vv=prepare_fan_purchase(schema)\n",
    "print(vv.shape)\n",
    "vv.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "generating gender count for venue ....\n",
      "(28, 4)\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>venue_name</th>\n",
       "      <th>gender</th>\n",
       "      <th>fan_id</th>\n",
       "      <th>collection_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Foundry</td>\n",
       "      <td>female</td>\n",
       "      <td>137</td>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Foundry</td>\n",
       "      <td>male</td>\n",
       "      <td>167</td>\n",
       "      <td>167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Fremantle Arts Centre</td>\n",
       "      <td>female</td>\n",
       "      <td>463</td>\n",
       "      <td>463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Fremantle Arts Centre</td>\n",
       "      <td>male</td>\n",
       "      <td>257</td>\n",
       "      <td>257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>HQ Complex</td>\n",
       "      <td>female</td>\n",
       "      <td>211</td>\n",
       "      <td>211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>HQ Complex</td>\n",
       "      <td>male</td>\n",
       "      <td>165</td>\n",
       "      <td>165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Hordern Pavilion</td>\n",
       "      <td>female</td>\n",
       "      <td>820</td>\n",
       "      <td>820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Hordern Pavilion</td>\n",
       "      <td>male</td>\n",
       "      <td>401</td>\n",
       "      <td>401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Hunter Lounge</td>\n",
       "      <td>female</td>\n",
       "      <td>156</td>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Hunter Lounge</td>\n",
       "      <td>male</td>\n",
       "      <td>121</td>\n",
       "      <td>121</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              venue_name  gender  fan_id  collection_id\n",
       "0                Foundry  female     137            137\n",
       "1                Foundry    male     167            167\n",
       "2  Fremantle Arts Centre  female     463            463\n",
       "3  Fremantle Arts Centre    male     257            257\n",
       "4             HQ Complex  female     211            211\n",
       "5             HQ Complex    male     165            165\n",
       "6       Hordern Pavilion  female     820            820\n",
       "7       Hordern Pavilion    male     401            401\n",
       "8          Hunter Lounge  female     156            156\n",
       "9          Hunter Lounge    male     121            121"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wc=create_venue_demopgrahics(schema)\n",
    "print(wc.shape)\n",
    "wc.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                              fan_id       venue_name  \\\n",
      "0  4ad6ef32a70b40231ed08bd14ee01beb606f2f34526067...  Melbourne Arena   \n",
      "1  08ea7600e75550f3e3d8a07a2063c55786ea6add985ca0...     UC Refectory   \n",
      "\n",
      "   time_in_seconds_to_venue  \n",
      "0                       NaN  \n",
      "1                   1348.35  \n"
     ]
    }
   ],
   "source": [
    "engine = get_rds_engine()\n",
    "schema = 'a441ffb172866cb4928c84a73de403ca15da4a54cc535e704413ab621' # unique customer schema id\n",
    "query = f\"\"\"\n",
    "    select distinct fan_id,venue_name, distance_driving_time as time_in_seconds_to_venue from {schema}.adhoc_address_distance\n",
    "        \"\"\"\n",
    "df = pd.read_sql(query, engine)\n",
    "df['time_in_seconds_to_venue'] = df['time_in_seconds_to_venue']/60\n",
    "print(df.head(2)) \n",
    "q=df[\"time_in_seconds_to_venue\"].quantile(0.90)\n",
    "df=df[df[\"time_in_seconds_to_venue\"] < q]\n",
    "p=df[\"time_in_seconds_to_venue\"].quantile(0.01)\n",
    "df=df[df[\"time_in_seconds_to_venue\"] >p]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0, 0.5, 'driving distance in minutes'), Text(0.5, 0, 'venue')]"
      ]
     },
     "execution_count": 5,
     "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": [
    "ax =sns.boxplot(x=\"venue_name\", y=\"time_in_seconds_to_venue\", data=df, showfliers = False)\n",
    "ax.set_xticklabels(ax.get_xticklabels(),rotation=90)\n",
    "ax.set(xlabel='venue', ylabel='driving distance in minutes')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>venue_name</th>\n",
       "      <th>fan_id</th>\n",
       "      <th>time_in_seconds_to_venue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Foundry</td>\n",
       "      <td>299</td>\n",
       "      <td>299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Fremantle Arts Centre</td>\n",
       "      <td>728</td>\n",
       "      <td>728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>HQ Complex</td>\n",
       "      <td>360</td>\n",
       "      <td>360</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Hordern Pavilion</td>\n",
       "      <td>88</td>\n",
       "      <td>88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Hunter Lounge</td>\n",
       "      <td>271</td>\n",
       "      <td>271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Melbourne Arena</td>\n",
       "      <td>55</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Metropolis Fremantle</td>\n",
       "      <td>413</td>\n",
       "      <td>413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Miami Marketta</td>\n",
       "      <td>1095</td>\n",
       "      <td>1095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Pittwater Park Rat Park</td>\n",
       "      <td>845</td>\n",
       "      <td>845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Studio</td>\n",
       "      <td>34</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>The Northern</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Thebarton Theatre</td>\n",
       "      <td>540</td>\n",
       "      <td>540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Tivoli</td>\n",
       "      <td>441</td>\n",
       "      <td>441</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>UC Refectory</td>\n",
       "      <td>694</td>\n",
       "      <td>694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Union Hall</td>\n",
       "      <td>303</td>\n",
       "      <td>303</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 venue_name  fan_id  time_in_seconds_to_venue\n",
       "0                   Foundry     299                       299\n",
       "1     Fremantle Arts Centre     728                       728\n",
       "2                HQ Complex     360                       360\n",
       "3          Hordern Pavilion      88                        88\n",
       "4             Hunter Lounge     271                       271\n",
       "5           Melbourne Arena      55                        55\n",
       "6      Metropolis Fremantle     413                       413\n",
       "7            Miami Marketta    1095                      1095\n",
       "8   Pittwater Park Rat Park     845                       845\n",
       "9                    Studio      34                        34\n",
       "10             The Northern     694                       694\n",
       "11        Thebarton Theatre     540                       540\n",
       "12                   Tivoli     441                       441\n",
       "13             UC Refectory     694                       694\n",
       "14               Union Hall     303                       303"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['venue_name']).count().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine = get_rds_engine()\n",
    "schema = 'a441ffb172866cb4928c84a73de403ca15da4a54cc535e704413ab620' # unique customer schema id\n",
    "\n",
    "query = f\"\"\"\n",
    "SELECT *  FROM {schema}.collection c \n",
    "where parent_id is not null\n",
    "\"\"\"\n",
    "\n",
    "df = pd.read_sql(query, engine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_colwidth', -1)\n",
    "df[['id','parent_id','name','status','source']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tableQuery =f\"\"\"\n",
    "SELECT *  FROM {schema}.collection_31_source cs\n",
    "\"\"\"\n",
    "collection1 = pd.read_sql(tableQuery, engine)\n",
    "print(collection1.shape)\n",
    "print(collection1.dtypes)\n",
    "print(collection1.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tableQuery10 =f\"\"\"\n",
    "SELECT * FROM information_schema.tables \n",
    "WHERE table_schema = 'a441ffb172866cb4928c84a73de403ca15da4a54cc535e704413ab620'\n",
    "\"\"\"\n",
    "collection10 = pd.read_sql(tableQuery10, engine)\n",
    "collection10[['table_name']].sort_values(by='table_name')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# print shapes for each original collection\n",
    "for count in [0,2,4,6,7,9,11,13,15,17,18,21,22,23,24,25,27,28,31,33]:\n",
    "    print('Collection: ' + str(count))\n",
    "    tableQueryX =f\"\"\"\n",
    "    SELECT *  FROM {schema}.collection_{count}_source cs\n",
    "    \"\"\"\n",
    "    collectionX = pd.read_sql(tableQueryX, engine)\n",
    "    print(collectionX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# print shapes for each enriched collection\n",
    "for count in [1,3,5,8,10,12,14,16,18,20,26,28,30,32,34]:\n",
    "    print('_Enriched collection: ' + str(count))\n",
    "    tableQueryX =f\"\"\"\n",
    "    SELECT *  FROM {schema}.collection_{count}_source cs\n",
    "    \"\"\"\n",
    "    collectionX = pd.read_sql(tableQueryX, engine)\n",
    "    print(collectionX.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "preparing dataset ...\n",
      "                                              fan_id  rfm_segment  \\\n",
      "0  c9a5774936f248be9ff14602cd623d8db2c3071b5727b0...  Hibernating   \n",
      "1  96a9bf266c1e4f95ae537b2d47784e2d484e22e00d981e...  Hibernating   \n",
      "\n",
      "   rfm_recency  rfm_frequency  rfm_monetary recency_value  frequency_value  \\\n",
      "0            1              2             1    2018-05-11              1.0   \n",
      "1            1              2             1    2018-05-13              1.0   \n",
      "\n",
      "   monetary_value   date_diff  \n",
      "0            45.0  761.418959  \n",
      "1            45.0  759.418959  \n",
      "fan_id                     object\n",
      "rfm_segment                object\n",
      "rfm_recency                 int64\n",
      "rfm_frequency               int64\n",
      "rfm_monetary                int64\n",
      "recency_value      datetime64[ns]\n",
      "frequency_value           float64\n",
      "monetary_value            float64\n",
      "date_diff                 float64\n",
      "dtype: object\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count    11415.000000\n",
       "mean       504.618823\n",
       "std        406.628573\n",
       "min          0.000000\n",
       "25%        146.418959\n",
       "50%        571.418959\n",
       "75%        761.418959\n",
       "max       1975.418959\n",
       "Name: date_diff, dtype: float64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "schema = 'a441ffb172866cb4928c84a73de403ca15da4a54cc535e704413ab621' # unique customer schema id\n",
    "df1 = prepare_rfm_dataset(schema, 'fan_purchase')\n",
    "print(df1.head(2))\n",
    "print(df1.dtypes)\n",
    "df1['date_diff'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fan_id                     object\n",
      "rfm_segment                object\n",
      "rfm_recency                 int64\n",
      "rfm_frequency               int64\n",
      "rfm_monetary                int64\n",
      "recency_value              object\n",
      "frequency_value           float64\n",
      "monetary_value            float64\n",
      "date_diff                 float64\n",
      "date               datetime64[ns]\n",
      "dtype: object\n",
      "        date recency_value  date_diff\n",
      "0 2018-05-11    2018-05-11   761.4058\n",
      "1 2018-05-13    2018-05-13   759.4058\n"
     ]
    }
   ],
   "source": [
    "# try to parse 2017-03-22 03:00:23+00 into date\n",
    "# rfmDF['date']=extract_datetime_part(rfmDF['recency_value'].str[:19], ['%Y-%m-%d %H:%M:%S'])\n",
    "#rfmDF[['date']]=pd.to_datetime(rfmDF['recency_value'].str[:19], format='%Y-%m-%d %H:%M:%S',errors='coerce')\n",
    "df1['date']=pd.to_datetime(df1['recency_value'], format='%Y-%m-%d',errors='coerce')\n",
    "df1['date_diff'] = (calculate_diff(df1['date'], pd.to_datetime('today'))) / pd.Timedelta(1, unit='d')\n",
    "print(df1.dtypes)\n",
    "print(df1[{'date','recency_value','date_diff'}].head(2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "collection3['Subtotal'] =pd.to_numeric(collection3['Subtotal'])\n",
    "collection3['userMonetary']=pd.to_numeric(collection3['userMonetary'])\n",
    "collection3['Shipping']=pd.to_numeric(collection3['Shipping'])\n",
    "collection3['Lineitem price']=pd.to_numeric(collection3['Lineitem price'])\n",
    "# collection3.dtypes\n",
    "collection3.groupby([\"Payment Method\"]).agg({'Lineitem sku':'count'}).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "collection3[collection3.fan_id=='00332bdbe9f330226382ca86a0b946ab03044d86290787dc73220c6f2fa8f97f']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "aggregatedData= collection3.groupby([\"fan_id\"]).agg({'Lineitem sku':'count','userMonetary':'sum','Subtotal':'sum','Shipping':'sum','Lineitem price':'sum'}).reset_index()\n",
    "print(aggregatedData.shape)\n",
    "print(aggregatedData.head())\n",
    "aggregatedData['Lineitem sku'].max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# shipped(2) and unshipped(4) orders- original files\n",
    "tableQueryS =f\"\"\"\n",
    "SELECT *  FROM {schema}.collection_4_source cs\n",
    "\"\"\"\n",
    "collectionS = pd.read_sql(tableQueryS, engine)\n",
    "collectionS[['userMonetary','userQuantity']] = collectionS[['userMonetary','userQuantity']].apply(pd.to_numeric, errors='coerce')\n",
    "print(collectionS.shape)\n",
    "print(collectionS.dtypes)\n",
    "print(collectionS.head())\n",
    "aggregatedDataS= collectionS.groupby([\"fan_id\"]).agg({'userMonetary':'sum','userQuantity':'sum'}).reset_index()\n",
    "print(aggregatedDataS.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "schema = \"bd345f915775993a4d3de1dae65b93b067ad69dde4286a1e1639e5cd2\"\n",
    "query2 = f\"\"\"\n",
    "SELECT venue_name, venue_address as fan_location\n",
    "FROM {schema}.event_data\n",
    ";\n",
    "\"\"\"\n",
    "\n",
    "df = pd.read_sql(query2, engine)\n",
    "\n",
    "dfObj = pd.DataFrame(columns=['venue_name', 'longitude', 'latitude'])\n",
    "\n",
    "for index, row in df.iterrows():\n",
    "    address = row['fan_location']\n",
    "    venue = row['venue_name']\n",
    "    # print(venue)\n",
    "    geolocation = get_geolocation(address)\n",
    "    df = parse_geolocation_to_df(geolocation)\n",
    "    # print(df[\"longitude\"])\n",
    "    # print(df[\"latitude\"].iloc[0])\n",
    "    dfObj = dfObj.append({'venue_name': venue, 'longitude': df[\"longitude\"].iloc[0], 'latitude': df[\"latitude\"].iloc[0]}, ignore_index=True)\n",
    "    print(\"All done!\")\n",
    "    \n",
    "dfObj.to_csv('ihw_venues_gps.csv', sep=',')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                              fan_id  cluster\n",
      "0  33353a380143896467524a892e1e64502aba647324ff56...        2\n",
      "1  2f8524b7f3c512dec961d3df9d95a4221c39d7db8350dc...        2\n",
      "2  ac6f22d3b3e2f1ef3a6fdd6b59ad50f4803a1a1505e8d1...        5\n",
      "3  ee28a32dc5597d75998773366882114ea25c99bac141b7...        2\n",
      "4  af527e3e563b2c8d7340562d36f97121dc7519b3265eec...        2\n",
      "(10123, 2)\n",
      "cluster\n",
      "1     522\n",
      "2    7136\n",
      "3      24\n",
      "4      67\n",
      "5    2374\n",
      "Name: fan_id, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# read ihw optin csv\n",
    "optinDf = pd.read_csv(\"merch_clusters.csv\")\n",
    "print(optinDf.head())\n",
    "print(optinDf.shape)\n",
    "clSize = optinDf.groupby(\"cluster\")[\"fan_id\"].count()\n",
    "print(clSize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "schema='bd345f915775993a4d3de1dae65b93b067ad69dde4286a1e1639e5cd2'\n",
    "\n",
    "optinDf.to_sql(f'merch_clusters',\n",
    "                                      engine,\n",
    "                                      schema=schema,\n",
    "                                      if_exists='replace'\n",
    "                                      # index_label='row_id',\n",
    "                                      # index=False\n",
    "                                      )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   male  female  total_tickets_per_fan  days_from_event  unique_events  \\\n",
      "0     1       0                      1              579              1   \n",
      "1     1       0                      1              579              1   \n",
      "2     0       1                      1              530              1   \n",
      "3     0       1                      1              884              1   \n",
      "4     0       1                      1              884              1   \n",
      "\n",
      "   distance  cluster  \n",
      "0   -1000.0        3  \n",
      "1   -1000.0        3  \n",
      "2   -1000.0        3  \n",
      "3   -1000.0        3  \n",
      "4   -1000.0        3  \n",
      "(2873, 7)\n",
      "male                       int64\n",
      "female                     int64\n",
      "total_tickets_per_fan      int64\n",
      "days_from_event            int64\n",
      "unique_events              int64\n",
      "distance                 float64\n",
      "cluster                    int64\n",
      "dtype: object\n",
      "['male', 'female', 'total_tickets_per_fan', 'days_from_event', 'unique_events', 'distance']\n"
     ]
    },
    {
     "ename": "KeyError",
     "evalue": "1229",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/indexes/base.py\u001b[0m in \u001b[0;36mget_loc\u001b[0;34m(self, key, method, tolerance)\u001b[0m\n\u001b[1;32m   2656\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2657\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2658\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 1229",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-3-4e7006681ae6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0moptinClusterDfFeat\u001b[0m \u001b[0;34m=\u001b[0m\u001b[0moptinClusterDf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdrop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'cluster'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moptinClusterDfFeat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtolist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 15\u001b[0;31m \u001b[0mfeature_importance_plot_location\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeature_importance\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrun_random_forest\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mclient_id\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0moptinClusterDf\u001b[0m \u001b[0;34m,\u001b[0m \u001b[0moptinClusterDfFeat\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtolist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtarget\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0malgo_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m<ipython-input-1-a8c3c7968ae1>\u001b[0m in \u001b[0;36mrun_random_forest\u001b[0;34m(client_id, df, cols, target, algo_name)\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m     \u001b[0mlogging\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minfo\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Running random forest to calculate feature importance ...\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 10\u001b[0;31m     \u001b[0mtrain_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mshuffle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     11\u001b[0m     \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcols\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m     \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_data\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mtarget\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/random.py\u001b[0m in \u001b[0;36mshuffle\u001b[0;34m(self, x, random)\u001b[0m\n\u001b[1;32m    273\u001b[0m                 \u001b[0;31m# pick an element in x[:i+1] with which to exchange x[i]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    274\u001b[0m                 \u001b[0mj\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrandbelow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 275\u001b[0;31m                 \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    276\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    277\u001b[0m             \u001b[0m_int\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/frame.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   2925\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnlevels\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2926\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_getitem_multilevel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2927\u001b[0;31m             \u001b[0mindexer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2928\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mis_integer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mindexer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2929\u001b[0m                 \u001b[0mindexer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mindexer\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/indexes/base.py\u001b[0m in \u001b[0;36mget_loc\u001b[0;34m(self, key, method, tolerance)\u001b[0m\n\u001b[1;32m   2657\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2658\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2659\u001b[0;31m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_cast_indexer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2660\u001b[0m         \u001b[0mindexer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget_indexer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmethod\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtolerance\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtolerance\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2661\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mindexer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0mindexer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msize\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 1229"
     ]
    }
   ],
   "source": [
    "# read ihw optin clusters with inputs for variable importance\n",
    "optinClusterDf = pd.read_csv(\"ihw_clusters_for_importance.csv\", sep=\",\")\n",
    "optinClusterDf.drop(columns=['gender', 'fan_id'],inplace=True)\n",
    "print(optinClusterDf.head())\n",
    "print(optinClusterDf.shape)\n",
    "print(optinClusterDf.dtypes)\n",
    "\n",
    "# df['cluster'] = optinClusterDf['cluster']\n",
    "#original_feature_vector['cluster'] = optinClusterDf['cluster']\n",
    "target = 'cluster'\n",
    "algo_name = 'Rkmeans'\n",
    "client_id = 'ihw'\n",
    "optinClusterDfFeat =optinClusterDf.drop(columns=['cluster'])\n",
    "print(optinClusterDfFeat.columns.tolist())\n",
    "feature_importance_plot_location, feature_importance = run_random_forest(client_id,optinClusterDf , optinClusterDfFeat.columns.tolist(), target, algo_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/psycopg2/__init__.py:144: UserWarning: The psycopg2 wheel package will be renamed from release 2.8; in order to keep installing from binary please use \"pip install psycopg2-binary\" instead. For details see: <http://initd.org/psycopg/docs/install.html#binary-install-from-pypi>.\n",
      "  \"\"\")\n"
     ]
    }
   ],
   "source": [
    "%run ./viz_utils.ipynb\n",
    "from IPython.display import Image\n",
    "\n",
    "# ihw\n",
    "schema='bd345f915775993a4d3de1dae65b93b067ad69dde4286a1e1639e5cd2'\n",
    "customer='IHW'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pixies\n",
    "schema='ace3c2bb4abd9fcaf9807975e6ab940131386dc9ecddb2b7e31288ff8'\n",
    "customer='Pixies'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_age_gender(schema,customer)\n",
    "plot_distance_from_venue(schema,customer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_fan_maps(schema, customer)\n",
    "# Image(filename=age_distro_plot_location)\n",
    "# Image(filename=gender_age_distro_plot_location)\n",
    "# Image(filename=gender_distro_plot_location)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_event_cluster_results(schema, customer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   cluster  size\n",
      "0        1   522\n",
      "1        2  7136\n",
      "2        3    24\n",
      "3        4    67\n",
      "4        5  2374\n"
     ]
    },
    {
     "ename": "DataError",
     "evalue": "No numeric types to aggregate",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mDataError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-3-4bddbb92f78d>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mplot_merch_cluster_results\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mschema\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustomer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m<ipython-input-2-a2fc4035c339>\u001b[0m in \u001b[0;36mplot_merch_cluster_results\u001b[0;34m(schema, customer)\u001b[0m\n\u001b[1;32m     63\u001b[0m     \u001b[0mclustersEventsMerchs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'average_item_value'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m\u001b[0mclustersEventsMerchs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'total_merch_value_per_fan'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mclustersEventsMerchs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'total_items_per_fan'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     64\u001b[0m     \u001b[0minput_columns\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m\"total_merch_value_per_fan\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\"total_items_per_fan\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\"average_item_value\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 65\u001b[0;31m     \u001b[0mclEvMeGr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mclustersEventsMerchs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgroupby\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"cluster\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mas_index\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0minput_columns\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmean\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mround\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdecimals\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     66\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mclEvMeGr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     67\u001b[0m     \u001b[0mdf_to_png\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mclEvMeGr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustomer\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'merch_cluster_means_table'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/groupby/groupby.py\u001b[0m in \u001b[0;36mmean\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1130\u001b[0m         \u001b[0mnv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalidate_groupby_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'mean'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m'numeric_only'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1131\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1132\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_cython_agg_general\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'mean'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1133\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mGroupByError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1134\u001b[0m             \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/groupby/generic.py\u001b[0m in \u001b[0;36m_cython_agg_general\u001b[0;34m(self, how, alt, numeric_only, min_count)\u001b[0m\n\u001b[1;32m     68\u001b[0m                             min_count=-1):\n\u001b[1;32m     69\u001b[0m         new_items, new_blocks = self._cython_agg_blocks(\n\u001b[0;32m---> 70\u001b[0;31m             how, alt=alt, numeric_only=numeric_only, min_count=min_count)\n\u001b[0m\u001b[1;32m     71\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_wrap_agged_blocks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_items\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnew_blocks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     72\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/python3/lib/python3.6/site-packages/pandas/core/groupby/generic.py\u001b[0m in \u001b[0;36m_cython_agg_blocks\u001b[0;34m(self, how, alt, numeric_only, min_count)\u001b[0m\n\u001b[1;32m    141\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    142\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_blocks\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 143\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mDataError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'No numeric types to aggregate'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    144\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    145\u001b[0m         \u001b[0;31m# reset the locs in the blocks to correspond to our\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mDataError\u001b[0m: No numeric types to aggregate"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_merch_cluster_results(schema, customer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " cluster       item group  count  proportion  min value  median value  max value\n",
      "       1          Upgrade    165        0.30      36.00          40.0     112.10\n",
      "       1          T-Shirt    104        0.19      10.00          50.0     100.00\n",
      "       1   Hooded Sweater     41        0.07      36.00          76.8     110.00\n",
      "       1  Long Sleeve Tee     32        0.06      35.95          70.0     104.85\n",
      "       1           Shorts     28        0.05      18.91          70.0      80.00\n",
      " cluster       item group  count  proportion  min value  median value  max value\n",
      "       2          T-Shirt   2601        0.36       5.00         28.95      42.00\n",
      "       2  Long Sleeve Tee    674        0.09      14.20         35.00      42.76\n",
      "       2              12\"    651        0.09      13.98         34.95      40.00\n",
      "       2          Singlet    534        0.07       4.95         25.00      39.95\n",
      "       2           Shorts    385        0.05       9.46         35.00      41.99\n",
      " cluster item group  count  proportion  min value  median value  max value\n",
      "       3    Upgrade     11        0.39      80.00         80.00     396.00\n",
      "       3        12\"      5        0.18      82.80        175.00     293.83\n",
      "       3    T-Shirt      3        0.11      23.95         60.00     119.80\n",
      "       3      Vinyl      3        0.11      11.99         24.99      39.99\n",
      "       3     Bundle      2        0.07     199.80        207.63     215.46\n",
      " cluster      item group  count  proportion  min value  median value  max value\n",
      "       4         Upgrade     33        0.49      118.0        198.00     298.00\n",
      "       4             12\"     11        0.16      120.0        144.00     250.00\n",
      "       4          Jacket     10        0.15      133.0        140.00     159.90\n",
      "       4  Hooded Sweater      7        0.10      114.9        119.90     209.85\n",
      "       4          Bundle      4        0.06      120.0        139.88     149.85\n",
      " cluster        item group  count  proportion  min value  median value  max value\n",
      "       5    Hooded Sweater    569        0.24      44.95          55.0      71.95\n",
      "       5            Jacket    381        0.16      45.58          70.0      99.95\n",
      "       5            Bundle    315        0.13      44.95          55.0      89.95\n",
      "       5               12\"    260        0.11      45.55          60.0      60.00\n",
      "       5  Crewneck Sweater    158        0.07      43.73          50.0      64.95\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_merch_cluster_top_items(schema, customer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "schema='bd345f915775993a4d3de1dae65b93b067ad69dde4286a1e1639e5cd2'\n",
    "customer='IHW'\n",
    "fan_venue_dist_plot_location=plot_distance_from_venue(schema,customer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# oa\n",
    "schema='a441ffb172866cb4928c84a73de403ca15da4a54cc535e704413ab621'\n",
    "customer = 'OA'\n",
    "\n",
    "age_distro_plot_location,gender_age_distro_plot_location,gender_distro_plot_location = plot_age_gender(schema,customer)\n",
    "#Image(filename=age_distro_plot_location)\n",
    "#Image(filename=gender_age_distro_plot_location)\n",
    "#Image(filename=gender_distro_plot_location)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_python3",
   "language": "python",
   "name": "conda_python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}