2019-08-06-bob-ross.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is siuba?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>cyl</th>\n",
       "      <th>avg_hp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>82.636364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>122.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>209.214286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   cyl      avg_hp\n",
       "0    4   82.636364\n",
       "1    6  122.285714\n",
       "2    8  209.214286"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from siuba import _, group_by, summarize, filter\n",
    "from siuba.data import mtcars\n",
    "\n",
    "(mtcars\n",
    "  >> group_by(_.cyl)\n",
    "  >> summarize(avg_hp = _.hp.mean())\n",
    "  )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "hp    146.6875\n",
       "dtype: float64"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mtcars.agg({'hp': lambda _: _.mean()})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>hp</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cyl</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>82.636364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>122.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>209.214286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             hp\n",
       "cyl            \n",
       "4     82.636364\n",
       "6    122.285714\n",
       "8    209.214286"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mtcars.groupby(['cyl']).agg({'hp': lambda _: _.mean()})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bob Ross Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from siuba import *\n",
    "from plotnine import *\n",
    "\n",
    "%load_ext blackcellmagic\n",
    "\n",
    "import matplotlib.cbook\n",
    "\n",
    "import warnings\n",
    "import plotnine\n",
    "warnings.filterwarnings(module='plotnine*', action='ignore')\n",
    "warnings.filterwarnings(module='matplotlib*', action='ignore')\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "bob_ross = pd.read_csv(\"https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2019/2019-08-06/bob-ross.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "tidy_bob = bob_ross >> gather(\"tag\", \"value\", _[\"APPLE_FRAME\":]) >> filter(_.value == 1)\n",
    "\n",
    "tidy_bob.columns = tidy_bob.columns.str.lower()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(403, 69)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bob_ross.shape"
   ]
  },
  {
   "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>episode</th>\n",
       "      <th>title</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>S10E09</td>\n",
       "      <td>\"MOUNTAIN OVAL\"</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>S02E04</td>\n",
       "      <td>\"SHADES OF GREY\"</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>S04E04</td>\n",
       "      <td>\"WINTER SAWSCAPE\"</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  episode              title   n\n",
       "0  S10E09    \"MOUNTAIN OVAL\"  15\n",
       "1  S02E04   \"SHADES OF GREY\"  14\n",
       "2  S04E04  \"WINTER SAWSCAPE\"  14"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_bob >> count(_.episode, _.title, sort = True) >> head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>tag</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>TREE</td>\n",
       "      <td>361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TREES</td>\n",
       "      <td>337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DECIDUOUS</td>\n",
       "      <td>227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CONIFER</td>\n",
       "      <td>212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CLOUDS</td>\n",
       "      <td>179</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         tag    n\n",
       "0       TREE  361\n",
       "1      TREES  337\n",
       "2  DECIDUOUS  227\n",
       "3    CONIFER  212\n",
       "4     CLOUDS  179"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_bob >> count(_.tag, sort = True) >> head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>value</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3221</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   value     n\n",
       "0      1  3221"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_bob >> count(_.value)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Episodes can have the same titles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>EPISODE</th>\n",
       "      <th>TITLE</th>\n",
       "      <th>APPLE_FRAME</th>\n",
       "      <th>AURORA_BOREALIS</th>\n",
       "      <th>BARN</th>\n",
       "      <th>BEACH</th>\n",
       "      <th>BOAT</th>\n",
       "      <th>BRIDGE</th>\n",
       "      <th>BUILDING</th>\n",
       "      <th>BUSHES</th>\n",
       "      <th>...</th>\n",
       "      <th>TOMB_FRAME</th>\n",
       "      <th>TREE</th>\n",
       "      <th>TREES</th>\n",
       "      <th>TRIPLE_FRAME</th>\n",
       "      <th>WATERFALL</th>\n",
       "      <th>WAVES</th>\n",
       "      <th>WINDMILL</th>\n",
       "      <th>WINDOW_FRAME</th>\n",
       "      <th>WINTER</th>\n",
       "      <th>WOOD_FRAMED</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>S08E02</td>\n",
       "      <td>\"LAKESIDE CABIN\"</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>S10E13</td>\n",
       "      <td>\"LAKESIDE CABIN\"</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    EPISODE             TITLE  APPLE_FRAME  AURORA_BOREALIS  BARN  BEACH  \\\n",
       "92   S08E02  \"LAKESIDE CABIN\"            0                0     0      0   \n",
       "129  S10E13  \"LAKESIDE CABIN\"            0                0     0      0   \n",
       "\n",
       "     BOAT  BRIDGE  BUILDING  BUSHES  ...  TOMB_FRAME  TREE  TREES  \\\n",
       "92      0       0         0       1  ...           0     1      1   \n",
       "129     0       0         0       0  ...           0     1      1   \n",
       "\n",
       "     TRIPLE_FRAME  WATERFALL  WAVES  WINDMILL  WINDOW_FRAME  WINTER  \\\n",
       "92              0          0      0         0             0       0   \n",
       "129             0          0      0         0             0       0   \n",
       "\n",
       "     WOOD_FRAMED  \n",
       "92             0  \n",
       "129            0  \n",
       "\n",
       "[2 rows x 69 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bob_ross >> filter(_.TITLE == '\"LAKESIDE CABIN\"')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Most common tags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (307313840)>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from siuba.dply.forcats import fct_reorder\n",
    "\n",
    "(tidy_bob\n",
    "    >> count(_.tag, sort=True)\n",
    "    >> mutate(tag=fct_reorder(_.tag, _.n))\n",
    "    >> filter(_.n > 39)\n",
    "    >> ggplot(aes(\"tag\", \"n\"))\n",
    "    + geom_col()\n",
    "    + coord_flip()\n",
    "    + labs(title=\"Frequency of Tags in Bob Ross Paintings\", x=\"Tag\", y=\"Number of Tags\")\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cond probability of trees (do trees have friends?)\n",
    "\n",
    "p(trees | tree) = p(tree AND trees) / p(tree)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tag</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>TREE</td>\n",
       "      <td>361</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>TREES</td>\n",
       "      <td>337</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      tag    n\n",
       "57   TREE  361\n",
       "58  TREES  337"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tidy_bob >> count(_.tag) >> filter(_.tag.isin([\"TREE\", \"TREES\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.93\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>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>337</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     n\n",
       "0  337"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(round(337 / 361, 2))\n",
    "\n",
    "bob_ross >> filter(_.TREE == 1, _.TREES == 1) >> count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Most common word in painting titles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (310343905)>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "title_word_count = (\n",
    "    bob_ross\n",
    "    >> select(_.EPISODE, _.TITLE)\n",
    "    >> mutate(word=_.TITLE.str.split(\" \"))\n",
    "    >> unnest(\"word\")\n",
    "    >> mutate(word=_.word.str.replace('\"', \"\"))\n",
    "    >> filter(~_.word.isin([\"THE\", \"OF\", \"IN\", \"AT\", \"A\"]))\n",
    "    >> count(_.word, sort=True)\n",
    ")\n",
    "\n",
    "(\n",
    "    title_word_count\n",
    "    >> head(15)\n",
    "    >> mutate(word=fct_reorder(_.word, _.n))\n",
    "    >> ggplot(aes(\"word\", \"n\"))\n",
    "    + geom_col()\n",
    "    + coord_flip()\n",
    "    + labs(title=\"Most Common Words in Painting Titles\", x=\"Word\", y=\"Number of Words\")\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tags grouped by season"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>tag</th>\n",
       "      <th>n</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>TREE</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TREES</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>DECIDUOUS</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CONIFER</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CLOUDS</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RIVER</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>LAKE</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>GRASS</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>MOUNTAIN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>BUSHES</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>CUMULUS</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>SNOWY_MOUNTAIN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               tag   n\n",
       "0             TREE  31\n",
       "1            TREES  31\n",
       "2        DECIDUOUS  18\n",
       "3          CONIFER  15\n",
       "4           CLOUDS  10\n",
       "5            RIVER   5\n",
       "6             LAKE   4\n",
       "7            GRASS   3\n",
       "8         MOUNTAIN   3\n",
       "9           BUSHES   2\n",
       "10         CUMULUS   2\n",
       "11  SNOWY_MOUNTAIN   1"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from siuba.dply.vector import row_number\n",
    "\n",
    "grouped_bob = tidy_bob >> mutate(\n",
    "    season_num=_.episode.str.extract(\"S([0-9]+)E([0-9]+)\").loc[:, 0].astype(int),\n",
    "    episode_num=_.episode.str.extract(\"S([0-9]+)E([0-9]+)\").loc[:, 1].astype(int),\n",
    ")\n",
    "\n",
    "(\n",
    "    grouped_bob\n",
    "    >> count(_.season_num, _.tag)\n",
    "    >> arrange(_.season_num, -_.n)\n",
    "    >> group_by(_.season_num)\n",
    "    >> mutate(rank = _.n.rank(method = \"dense\", ascending = False).astype(int))\n",
    "    >> filter(_[\"rank\"] < 4)\n",
    "    >> ungroup()\n",
    "    >> count(_.tag, sort = True)\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "formats": "ipynb,Rmd//Rmd"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "165px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}