mariusvniekerk/Download_stats.ipynb

## Download_stats.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conda-Forge Download Statistics\n",
    "\n",
    "This notebook demonstrates how to load and use Conda-forge\n",
    "package data.  For more details, see the [Github repository](https://github.com/ContinuumIO/anaconda-package-data/blob/master/README.md).  Due to limitations on Binder, you might find some of the analysis examples below run slowly or require more memory than is available on the Binder instance.  Feel free to download this notebook locally and run it.\n",
    "\n",
    "\n",
    "## Setting up\n",
    "\n",
    "To start we need to install the needed packages by running `conda install dask intake numpy pandas` and `conda install -c conda-forge hvplot`. Then we can import the packages:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !conda install -c conda-forge --yes hvplot intake intake-parquet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dask.dataframe as dd\n",
    "from datetime import datetime\n",
    "import hvplot.pandas\n",
    "import intake\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import distributed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "client = distributed.Client()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading Data\n",
    "\n",
    "There are multiple ways to load Anaconda package data. Below we show examples of loading one month of data for December 2018.\n",
    "\n",
    "#### Method 1:  load data from S3 url\n",
    "\n",
    "First, we can read parquet files directly from S3 url. We recommend using `dask.dataframe` to read data files into a Dask DataFrame. Please visit the [Dask website](http://docs.dask.org/en/latest/dataframe.html) for more information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = dd.read_parquet(\n",
    "    's3://anaconda-package-data/conda/hourly/*/*/*.parquet',\n",
    "    columns=['time', 'data_source', 'counts', 'pkg_name', 'pkg_python'],\n",
    "    filters=[('data_source', '=', 'conda-forge')],\n",
    "    storage_options={'anon': True}\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filter out non-conda-forge\n",
    "df_conda_forge_only = df[df.data_source == 'conda-forge']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><strong>Dask DataFrame Structure:</strong></div>\n",
       "<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>time</th>\n",
       "      <th>data_source</th>\n",
       "      <th>counts</th>\n",
       "      <th>pkg_name</th>\n",
       "      <th>pkg_python</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>npartitions=877</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <td>datetime64[ns]</td>\n",
       "      <td>category[unknown]</td>\n",
       "      <td>int64</td>\n",
       "      <td>category[unknown]</td>\n",
       "      <td>category[unknown]</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>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</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></th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "<div>Dask Name: getitem, 3508 tasks</div>"
      ],
      "text/plain": [
       "Dask DataFrame Structure:\n",
       "                           time        data_source counts           pkg_name         pkg_python\n",
       "npartitions=877                                                                                \n",
       "                 datetime64[ns]  category[unknown]  int64  category[unknown]  category[unknown]\n",
       "                            ...                ...    ...                ...                ...\n",
       "...                         ...                ...    ...                ...                ...\n",
       "                            ...                ...    ...                ...                ...\n",
       "                            ...                ...    ...                ...                ...\n",
       "Dask Name: getitem, 3508 tasks"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_conda_forge_only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_period_py = df_conda_forge_only.groupby(\n",
    "    [df_conda_forge_only.time.dt.to_period('D'), df_conda_forge_only.pkg_python]\n",
    ").counts.sum().compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "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>pkg_python</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-03-29</th>\n",
       "      <td></td>\n",
       "      <td>41032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-29</th>\n",
       "      <td>2.7</td>\n",
       "      <td>32246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-29</th>\n",
       "      <td>3.4</td>\n",
       "      <td>2381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-29</th>\n",
       "      <td>3.5</td>\n",
       "      <td>25041</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-29</th>\n",
       "      <td>3.6</td>\n",
       "      <td>20409</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           pkg_python  counts\n",
       "time                         \n",
       "2017-03-29              41032\n",
       "2017-03-29        2.7   32246\n",
       "2017-03-29        3.4    2381\n",
       "2017-03-29        3.5   25041\n",
       "2017-03-29        3.6   20409"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_period_py.reset_index(level=1).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Overall growth rate of conda-forge by downloads"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "daily = df_period_py.reset_index(level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "quarterly = daily.groupby(daily.index.to_series().dt.asfreq('Q')).counts.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "quarterly_yoy_growth = (quarterly - quarterly.shift(4)) / quarterly.shift(4) * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "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>downloads</th>\n",
       "      <th>yoy_growth_perc</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017Q1</th>\n",
       "      <td>701645</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017Q2</th>\n",
       "      <td>23172007</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017Q3</th>\n",
       "      <td>34139060</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017Q4</th>\n",
       "      <td>38632180</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018Q1</th>\n",
       "      <td>56972415</td>\n",
       "      <td>8019.834817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018Q2</th>\n",
       "      <td>64620212</td>\n",
       "      <td>178.871882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018Q3</th>\n",
       "      <td>98377527</td>\n",
       "      <td>188.167064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018Q4</th>\n",
       "      <td>140307194</td>\n",
       "      <td>263.187358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019Q1</th>\n",
       "      <td>145704942</td>\n",
       "      <td>155.746473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019Q2</th>\n",
       "      <td>115412939</td>\n",
       "      <td>78.601919</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        downloads  yoy_growth_perc\n",
       "time                              \n",
       "2017Q1     701645              NaN\n",
       "2017Q2   23172007              NaN\n",
       "2017Q3   34139060              NaN\n",
       "2017Q4   38632180              NaN\n",
       "2018Q1   56972415      8019.834817\n",
       "2018Q2   64620212       178.871882\n",
       "2018Q3   98377527       188.167064\n",
       "2018Q4  140307194       263.187358\n",
       "2019Q1  145704942       155.746473\n",
       "2019Q2  115412939        78.601919"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame({\"downloads\": quarterly, \"yoy_growth_perc\": quarterly_yoy_growth})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "unstacked = df_period_py.unstack()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Give this a nicer name\n",
    "unstacked.columns = [\"non-python\"] + list(unstacked.columns[1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "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>non-python</th>\n",
       "      <th>2.6</th>\n",
       "      <th>2.7</th>\n",
       "      <th>3.4</th>\n",
       "      <th>3.5</th>\n",
       "      <th>3.6</th>\n",
       "      <th>3.7</th>\n",
       "      <th>3.8</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2017-03-29</th>\n",
       "      <td>41032.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>32246.0</td>\n",
       "      <td>2381.0</td>\n",
       "      <td>25041.0</td>\n",
       "      <td>20409.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-30</th>\n",
       "      <td>103080.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>72273.0</td>\n",
       "      <td>7785.0</td>\n",
       "      <td>60248.0</td>\n",
       "      <td>52200.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-03-31</th>\n",
       "      <td>92143.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>70491.0</td>\n",
       "      <td>9720.0</td>\n",
       "      <td>60124.0</td>\n",
       "      <td>52466.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-01</th>\n",
       "      <td>40828.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30976.0</td>\n",
       "      <td>2921.0</td>\n",
       "      <td>27220.0</td>\n",
       "      <td>23451.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2017-04-02</th>\n",
       "      <td>43776.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>30651.0</td>\n",
       "      <td>2568.0</td>\n",
       "      <td>26186.0</td>\n",
       "      <td>22949.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            non-python  2.6      2.7     3.4      3.5      3.6  3.7  3.8\n",
       "time                                                                    \n",
       "2017-03-29     41032.0  NaN  32246.0  2381.0  25041.0  20409.0  NaN  NaN\n",
       "2017-03-30    103080.0  NaN  72273.0  7785.0  60248.0  52200.0  NaN  NaN\n",
       "2017-03-31     92143.0  6.0  70491.0  9720.0  60124.0  52466.0  NaN  NaN\n",
       "2017-04-01     40828.0  NaN  30976.0  2921.0  27220.0  23451.0  NaN  NaN\n",
       "2017-04-02     43776.0  NaN  30651.0  2568.0  26186.0  22949.0  NaN  NaN"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "unstacked.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f6b1a512860>"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "unstacked.asfreq('M').stack().groupby(level=0).sum().plot.line(\n",
    "    figsize=(16,9),\n",
    "    title=\"Total packages downloaded by month, by python version requirement\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f6af7056048>"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "unstacked.groupby(unstacked.asfreq('M').index).sum().plot.line(\n",
    "    figsize=(16,9),\n",
    "    title=\"Packages downloaded by month, by python version requirement\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Most popular packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "total_package_downloads = df_conda_forge_only.groupby(df_conda_forge_only.pkg_name).counts.sum().compute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "pkg_downloads = total_package_downloads[total_package_downloads > 0].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "pkg_perc = pkg_downloads / pkg_downloads.sum() * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>downloads</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pkg_name</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>certifi</th>\n",
       "      <td>12626379</td>\n",
       "      <td>1.758450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>openssl</th>\n",
       "      <td>11193308</td>\n",
       "      <td>1.558869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ca-certificates</th>\n",
       "      <td>10456441</td>\n",
       "      <td>1.456247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conda</th>\n",
       "      <td>7711862</td>\n",
       "      <td>1.074015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>python</th>\n",
       "      <td>7571981</td>\n",
       "      <td>1.054535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>setuptools</th>\n",
       "      <td>6797022</td>\n",
       "      <td>0.946608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>zlib</th>\n",
       "      <td>6366129</td>\n",
       "      <td>0.886598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pip</th>\n",
       "      <td>6334639</td>\n",
       "      <td>0.882212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sqlite</th>\n",
       "      <td>6165625</td>\n",
       "      <td>0.858674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wheel</th>\n",
       "      <td>6153064</td>\n",
       "      <td>0.856925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tk</th>\n",
       "      <td>6076110</td>\n",
       "      <td>0.846208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pytz</th>\n",
       "      <td>5685349</td>\n",
       "      <td>0.791787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>six</th>\n",
       "      <td>5662464</td>\n",
       "      <td>0.788600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>python-dateutil</th>\n",
       "      <td>5565805</td>\n",
       "      <td>0.775138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>libpng</th>\n",
       "      <td>5334109</td>\n",
       "      <td>0.742871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>jpeg</th>\n",
       "      <td>5324537</td>\n",
       "      <td>0.741538</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>numpy</th>\n",
       "      <td>5319361</td>\n",
       "      <td>0.740817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>icu</th>\n",
       "      <td>5256169</td>\n",
       "      <td>0.732016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freetype</th>\n",
       "      <td>5169753</td>\n",
       "      <td>0.719981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>openblas</th>\n",
       "      <td>5148201</td>\n",
       "      <td>0.716980</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 downloads  percentage\n",
       "pkg_name                              \n",
       "certifi           12626379    1.758450\n",
       "openssl           11193308    1.558869\n",
       "ca-certificates   10456441    1.456247\n",
       "conda              7711862    1.074015\n",
       "python             7571981    1.054535\n",
       "setuptools         6797022    0.946608\n",
       "zlib               6366129    0.886598\n",
       "pip                6334639    0.882212\n",
       "sqlite             6165625    0.858674\n",
       "wheel              6153064    0.856925\n",
       "tk                 6076110    0.846208\n",
       "pytz               5685349    0.791787\n",
       "six                5662464    0.788600\n",
       "python-dateutil    5565805    0.775138\n",
       "libpng             5334109    0.742871\n",
       "jpeg               5324537    0.741538\n",
       "numpy              5319361    0.740817\n",
       "icu                5256169    0.732016\n",
       "freetype           5169753    0.719981\n",
       "openblas           5148201    0.716980"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame({'downloads': pkg_downloads, 'percentage': pkg_perc}).head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Total number of downloads"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "718040121"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gdf = df_conda_forge_only.groupby(df_conda_forge_only.time)\n",
    "out_s = gdf.counts.sum()\n",
    "total_downloads = out_s.compute()\n",
    "sum(total_downloads)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "py36 mvn",
   "language": "python",
   "name": "py36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}