{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np \n",
    "import seaborn as sns \n",
    "import matplotlib.pylab as plt \n",
    "import sketch\n",
    "#plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "School Name    object\n",
       "School Uuid    object\n",
       "Customer Id    object\n",
       "Path           object\n",
       "Apr 7 - 13      int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.set_option('display.max_rows', 100)\n",
    "pageviews = pd.read_csv(\"/Users/normrasmussen/Documents/Work/Data/GraphQuery (4).csv\", index_col=None)\n",
    "pageviews.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>academy</th>\n",
       "      <th>academy_id</th>\n",
       "      <th>org_id</th>\n",
       "      <th>path</th>\n",
       "      <th>apr_week</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Compass Academy</td>\n",
       "      <td>dc62bd23-e093-49ff-9f85-b8e3d6dc9e34</td>\n",
       "      <td>49efbc2d-004a-4130-819a-2f5cdd59413f</td>\n",
       "      <td>/people</td>\n",
       "      <td>356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CHEK Institute</td>\n",
       "      <td>c2adf18f-9247-403f-9914-f9ce7d412d17</td>\n",
       "      <td>4d7e5559-a90f-4c1c-9de7-88d4495ae76a</td>\n",
       "      <td>/people</td>\n",
       "      <td>288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GenTech Associates Academy</td>\n",
       "      <td>d38a8181-f864-40cd-91da-b47b05f8bf69</td>\n",
       "      <td>8fe4e045-0882-4a7d-a695-1477f1aa32d7</td>\n",
       "      <td>/people</td>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CTAA&amp;#39;s School</td>\n",
       "      <td>e79c2971-f7dd-4f7b-93b3-c3192c00824f</td>\n",
       "      <td>677b83a3-20d8-4073-9cfb-6a28c6d53b92</td>\n",
       "      <td>/people</td>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Solo Internal Training</td>\n",
       "      <td>72047a94-38c4-460e-90e2-2a3df5096852</td>\n",
       "      <td>c05d6a25-cc80-447a-9b85-7b222ad9e4fd</td>\n",
       "      <td>/media</td>\n",
       "      <td>129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>Lighthouse by LJ Hooker Group</td>\n",
       "      <td>cb56a39f-b6a1-4c0a-893f-9e0e93fe6c24</td>\n",
       "      <td>7ccf4a79-c896-4865-95e4-dfe134119943</td>\n",
       "      <td>/instructor_led_trainings</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>Mark43 University</td>\n",
       "      <td>85e2acae-6a0b-4f1c-a3e1-dd113218321d</td>\n",
       "      <td>ea5c9708-73f9-4905-a759-7b68d0128206</td>\n",
       "      <td>/courses</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>Mizuno Academy</td>\n",
       "      <td>f61cd40c-6cbf-41b3-a797-78b923137c23</td>\n",
       "      <td>b99abead-4af0-487d-b8a4-d77b2c8dded4</td>\n",
       "      <td>/communications/emails/broadcast</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>Capstone Scholars Program</td>\n",
       "      <td>66f38581-27d9-4d9d-8134-edd09c87940a</td>\n",
       "      <td>abbe2201-8500-4067-aeaf-73fa99824784</td>\n",
       "      <td>/people</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>Garrett Virtual Academy</td>\n",
       "      <td>4f211014-bf83-4c8d-8779-d743777ca8cf</td>\n",
       "      <td>26370510-3f8a-4d3a-9808-3d7ba2388e97</td>\n",
       "      <td>/people/invite</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>123 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                           academy                            academy_id   \n",
       "0                 Compass Academy   dc62bd23-e093-49ff-9f85-b8e3d6dc9e34  \\\n",
       "1                   CHEK Institute  c2adf18f-9247-403f-9914-f9ce7d412d17   \n",
       "2       GenTech Associates Academy  d38a8181-f864-40cd-91da-b47b05f8bf69   \n",
       "3                CTAA&#39;s School  e79c2971-f7dd-4f7b-93b3-c3192c00824f   \n",
       "4           Solo Internal Training  72047a94-38c4-460e-90e2-2a3df5096852   \n",
       "..                             ...                                   ...   \n",
       "118  Lighthouse by LJ Hooker Group  cb56a39f-b6a1-4c0a-893f-9e0e93fe6c24   \n",
       "119              Mark43 University  85e2acae-6a0b-4f1c-a3e1-dd113218321d   \n",
       "120                 Mizuno Academy  f61cd40c-6cbf-41b3-a797-78b923137c23   \n",
       "121     Capstone Scholars Program   66f38581-27d9-4d9d-8134-edd09c87940a   \n",
       "122        Garrett Virtual Academy  4f211014-bf83-4c8d-8779-d743777ca8cf   \n",
       "\n",
       "                                   org_id                              path   \n",
       "0    49efbc2d-004a-4130-819a-2f5cdd59413f                           /people  \\\n",
       "1    4d7e5559-a90f-4c1c-9de7-88d4495ae76a                           /people   \n",
       "2    8fe4e045-0882-4a7d-a695-1477f1aa32d7                           /people   \n",
       "3    677b83a3-20d8-4073-9cfb-6a28c6d53b92                           /people   \n",
       "4    c05d6a25-cc80-447a-9b85-7b222ad9e4fd                            /media   \n",
       "..                                    ...                               ...   \n",
       "118  7ccf4a79-c896-4865-95e4-dfe134119943         /instructor_led_trainings   \n",
       "119  ea5c9708-73f9-4905-a759-7b68d0128206                          /courses   \n",
       "120  b99abead-4af0-487d-b8a4-d77b2c8dded4  /communications/emails/broadcast   \n",
       "121  abbe2201-8500-4067-aeaf-73fa99824784                           /people   \n",
       "122  26370510-3f8a-4d3a-9808-3d7ba2388e97                    /people/invite   \n",
       "\n",
       "     apr_week  \n",
       "0         356  \n",
       "1         288  \n",
       "2         137  \n",
       "3         136  \n",
       "4         129  \n",
       "..        ...  \n",
       "118        32  \n",
       "119        32  \n",
       "120        32  \n",
       "121        32  \n",
       "122        32  \n",
       "\n",
       "[123 rows x 5 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pageviews = pageviews.rename(columns={\n",
    "    'School Name':'academy',\n",
    "    'School Uuid':'academy_id',\n",
    "    'Customer Id':'org_id',\n",
    "    'Path':'path',\n",
    "    'Apr 7 - 13': 'apr_week',\n",
    "})\n",
    "pageviews.head(123)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pageviews.apr_week = pageviews.apr_week.astype(int)\n",
    "# pageviews.academy_id = pageviews.academy_id.astype('category').cat.codes\n",
    "# pageviews.org_id = pageviews.org_id.astype('category').cat.codes\n",
    "# pageviews['academy']=pageviews['academy'].astype('category').cat.codes\n",
    "# pageviews['path']=pageviews['path'].astype('category').cat.codes\n",
    "#  pageviews.plot(kind='scatter', x='apr_week', y='path')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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></th>\n",
       "      <th>apr_week</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>academy</th>\n",
       "      <th>path</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Vivvix University for Ad Intel</th>\n",
       "      <th>/account_links</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/courses</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"8\" valign=\"top\">'@properties Academy</th>\n",
       "      <th>/account_links</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/courses</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/courses</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/edit</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/memberships</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/memberships/new</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/instructor_led_trainings</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                    apr_week\n",
       "academy                         path                                                        \n",
       " Vivvix University for Ad Intel /account_links                                             1\n",
       "                                /courses                                                   1\n",
       "'@properties Academy            /account_links                                             8\n",
       "                                /courses                                                  11\n",
       "                                /groups                                                    4\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/co...         1\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/edit          1\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/me...         8\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/me...         3\n",
       "                                /instructor_led_trainings                                 11"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#pageviews = pageviews.groupby(['academy', 'path']).agg({'apr_week':'sum'})\n",
    "#pageviews.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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></th>\n",
       "      <th>apr_week</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>academy</th>\n",
       "      <th>path</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Vivvix University for Ad Intel</th>\n",
       "      <th>/account_links</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/courses</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"8\" valign=\"top\">'@properties Academy</th>\n",
       "      <th>/account_links</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/courses</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/courses</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/edit</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/memberships</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/memberships/new</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>/instructor_led_trainings</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                    apr_week\n",
       "academy                         path                                                        \n",
       " Vivvix University for Ad Intel /account_links                                             1\n",
       "                                /courses                                                   1\n",
       "'@properties Academy            /account_links                                             8\n",
       "                                /courses                                                  11\n",
       "                                /groups                                                    4\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/co...         1\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/edit          1\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/me...         8\n",
       "                                /groups/2a09819b-a19a-4e87-91c6-8b50ee3c46f8/me...         3\n",
       "                                /instructor_led_trainings                                 11"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pageviews.reset_index\n",
    "pageviews.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         True\n",
       "1         True\n",
       "2         True\n",
       "3         True\n",
       "4        False\n",
       "         ...  \n",
       "13861     True\n",
       "13862     True\n",
       "13863    False\n",
       "13864    False\n",
       "13865    False\n",
       "Name: path, Length: 13866, dtype: bool"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "startswith_list = ('/courses', '/people', 'learning_paths', '/instructor_led_trainings', '/preview', '/groups', '/account_links')\n",
    "#for root in startswith_list:\n",
    "#    pageviews.path.str.startswith(root, na=False)\n",
    "    #pageviews['root'] = pageviews.path.apply(lambda x: root if pageviews.path.str.startswith(root, na=False) else 'None')\n",
    "\n",
    "pageviews.path.str.startswith(startswith_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "jupyter",
   "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": "3.11.1"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}