MLtG.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine Learning: the Gathering\n",
    "\n",
    "In Magic: the Gathering, a collectable card game competitive players tend to gravitate towards a few dozen of the best decks made up out of a subset of all available cards. For instance in the Legacy format nearly all 18000 cards can be played, yet you'll only see about 500 show up in tournaments with some (e.g. Brainstorm and Force of Will) showing up in > 50% of all high-ranking decks.\n",
    "\n",
    "When playing in such an event it is key to quickly identify your opponent's deck and adapt your own game plan accordingly. Top players are able to very quickly do this, bad ones like myself need a few more turns. Here we'll see if we can train a model that takes in a few known cards and outputs a prediction which deck is being played.\n",
    "\n",
    "This notebook consists of a few sections:\n",
    "\n",
    "  * Downloading the data (requests and Beautiful Soup)\n",
    "  * Checking the data (pandas, seaborn)\n",
    "  * Building a classifier (sklearn)\n",
    "  * Balancing the dataset (imbalanced-learn, sklearn) and retraining a classifier\n",
    "  * Running the model\n",
    "\n",
    "\n",
    "## Step 01: getting some decklists\n",
    "\n",
    "We'll get decklists from MTG Top 8 a website that collects decklists from the best 8 players in different tournaments. We'll focus on the Legacy format, using only decks from the last two weeks ( https://www.mtgtop8.com/format?f=LE&meta=34 ). Using Beautiful Soup we'll get all decks lists from the different archetypes.\n",
    "\n",
    "This process is rather straightforward, we get the website using requests, pick up the links we want using Beautiful Soup, follow that links using requests, pick up the correct link again using Beautiful Soup and finally get the decklis using requests again. Parsing HTML tends to be tedious and if the website updates this will likely break.\n",
    "\n",
    "You can pick up at Step 02 if you don't want to download these files. The results (from 27/07/2020) are included in this repository."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load all libraries\n",
    "\n",
    "from bs4 import BeautifulSoup\n",
    "import requests\n",
    "import re\n",
    "from tqdm import tqdm\n",
    "import pandas as pd\n",
    "from collections import Counter\n",
    "\n",
    "from scipy.spatial.distance import pdist, squareform\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "base_url = \"https://www.mtgtop8.com/\"\n",
    "initial_url = base_url + \"format?f=LE&meta=34\"\n",
    "deck_url = 'https://www.mtgtop8.com/mtgo?d=%s'\n",
    "\n",
    "resp = requests.get(initial_url)\n",
    "soup = BeautifulSoup(resp.text, 'html.parser')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'UR Aggro': 'archetype?a=213&meta=34&f=LE', 'Dragon Stompy': 'archetype?a=78&meta=34&f=LE', 'Death & Taxes': 'archetype?a=319&meta=34&f=LE', \"Death's Shadow\": 'archetype?a=772&meta=34&f=LE', 'Burn': 'archetype?a=20&meta=34&f=LE', 'Goblins': 'archetype?a=18&meta=34&f=LE', 'Merfolk': 'archetype?a=19&meta=34&f=LE', 'Canadian Threshold': 'archetype?a=23&meta=34&f=LE', 'Mono Black Aggro': 'archetype?a=237&meta=34&f=LE', 'Jund': 'archetype?a=301&meta=34&f=LE', 'BUG Midrange': 'archetype?a=86&meta=34&f=LE', 'Maverick': 'archetype?a=128&meta=34&f=LE', 'Ninja': 'archetype?a=898&meta=34&f=LE', 'Delver (Other)': 'archetype?a=149&meta=34&f=LE', 'Hollow One Madness': 'archetype?a=996&meta=34&f=LE', 'Other - Aggro': 'archetype?a=56&meta=34&f=LE', '4c Control': 'archetype?a=766&meta=34&f=LE', 'Artifacts Blue': 'archetype?a=473&meta=34&f=LE', 'Lands': 'archetype?a=36&meta=34&f=LE', 'Stoneblade': 'archetype?a=172&meta=34&f=LE', 'UWx Control': 'archetype?a=242&meta=34&f=LE', 'Bant Control': 'archetype?a=883&meta=34&f=LE', 'BUG Control': 'archetype?a=178&meta=34&f=LE', 'Grixis Control': 'archetype?a=478&meta=34&f=LE', 'Other - Control': 'archetype?a=61&meta=34&f=LE', 'Dark Depths': 'archetype?a=555&meta=34&f=LE', 'Reanimator': 'archetype?a=32&meta=34&f=LE', 'Storm': 'archetype?a=35&meta=34&f=LE', 'Elves': 'archetype?a=38&meta=34&f=LE', 'Doomsday': 'archetype?a=161&meta=34&f=LE', 'Show and Tell': 'archetype?a=33&meta=34&f=LE', 'Painter': 'archetype?a=87&meta=34&f=LE', 'Aluren': 'archetype?a=82&meta=34&f=LE', 'Cephalid Breakfast': 'archetype?a=96&meta=34&f=LE', 'Infect': 'archetype?a=98&meta=34&f=LE', 'All my Spells': 'archetype?a=385&meta=34&f=LE', 'Enchantress': 'archetype?a=91&meta=34&f=LE', 'Cloudpost Ramp': 'archetype?a=376&meta=34&f=LE', 'Hive Mind': 'archetype?a=162&meta=34&f=LE', 'Other - Combo': 'archetype?a=66&meta=34&f=LE'}\n"
     ]
    }
   ],
   "source": [
    "archetypes = {}\n",
    "decks = {}\n",
    "\n",
    "# First find all archetypes and the link to their details\n",
    "for a in soup.find_all('a'):\n",
    "    if \"archetype\" in a.get('href', ''):\n",
    "        archetypes[a.string] = a['href']\n",
    "        \n",
    "print(archetypes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████| 40/40 [00:45<00:00,  1.14s/it]\n"
     ]
    }
   ],
   "source": [
    "pattern = re.compile(\"d=(\\d+)\")  # Regex to extract deck ID from url\n",
    "\n",
    "# For each archetype get the deck IDs\n",
    "for archetype, url in tqdm(archetypes.items()):\n",
    "    current_url = base_url + url\n",
    "    resp = requests.get(current_url)\n",
    "    current_soup = BeautifulSoup(resp.text, 'html.parser')\n",
    "    \n",
    "    deck_urls = [a.get('href') for a in current_soup.find_all('a') if 'event' in a.get('href', '') and 'd=' in a.get('href', '')] \n",
    "    decks[archetype] =  [pattern.search(d).group(1) for d in deck_urls]\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████| 40/40 [02:39<00:00,  3.98s/it]\n"
     ]
    }
   ],
   "source": [
    "# Fetch decklist\n",
    "data = []\n",
    "\n",
    "for archetype, deck_ids in tqdm(decks.items()):\n",
    "    for i, deck_id in enumerate(deck_ids, start=1):\n",
    "        current_url = deck_url % (deck_id)\n",
    "        resp = requests.get(current_url)\n",
    "\n",
    "        for line in resp.text.split(\"\\r\\n\"):\n",
    "            # We don't need the sideboard, break once this line is found\n",
    "            if line.startswith(\"Sideboard\"):\n",
    "                break\n",
    "            card_count, card_name = line.split(\" \", 1)\n",
    "            data.append({\n",
    "                \"archetype\": archetype,\n",
    "                \"deck\": archetype + ' ' + str(i),\n",
    "                \"card\": card_name,\n",
    "                \"count\": int(card_count)\n",
    "            })\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 02: Data Science !\n",
    "\n",
    "Now we got for each archetype all decks and all the main deck cards for that deck in a list of dictionaries, this can be directly converted into a pandas dataframe. Using the .pivot function we'll be able to convert that table into a count matrix.\n",
    "\n",
    "We'll first prepare the dataframes and store them as csv files (comma separated values) so we can pick up from this point without downloading all decklists again."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data)\n",
    "card_table = df.pivot(index='deck', columns='card', values='count').fillna(0)\n",
    "archetype_df = df[[\"archetype\", \"deck\"]].drop_duplicates().set_index(\"deck\").reindex(card_table.index)\n",
    "\n",
    "df.to_csv('decks_table.csv')\n",
    "card_table.to_csv('card_table.csv')\n",
    "archetype_df.to_csv('archetypes.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If you don't want to download the decklists yourself, start from this point \n",
    "\n",
    "df = pd.read_csv('decks_table.csv', index_col=0)\n",
    "card_table = pd.read_csv('card_table.csv', index_col=0)\n",
    "archetype_df = pd.read_csv('archetypes.csv', index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 39 archetypes, the most popular ones are: 4c Control(20), UR Aggro(20), Dragon Stompy(12), Artifacts Blue(11), Death & Taxes(11)\n",
      "The 10 most popular cards:\n",
      "Force of Will           372.0\n",
      "Brainstorm              355.0\n",
      "Ponder                  339.0\n",
      "Wasteland               261.0\n",
      "Misty Rainforest        237.0\n",
      "Swords to Plowshares    194.0\n",
      "Plains                  185.0\n",
      "Polluted Delta          175.0\n",
      "Scalding Tarn           163.0\n",
      "Ancient Tomb            159.0\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Let's do some sanity checks first\n",
    "\n",
    "arch_counter = Counter(archetype_df['archetype'])\n",
    "most_common = [a + '(' + str(c) + ')' for a,c in arch_counter.most_common(5)]\n",
    "print(\"There are %d archetypes, the most popular ones are: %s\" % (len(arch_counter), ', '.join(most_common)))\n",
    "\n",
    "print(\"The 10 most popular cards:\")\n",
    "print(card_table.sum().sort_values(ascending=False).head(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\u0123248\\AppData\\Local\\Temp\\ipykernel_4752\\1734371086.py:3: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df_mean = df.groupby(['archetype', 'card'], as_index=False).mean()\n",
      "c:\\GBW_MyPrograms\\condaenv\\lib\\site-packages\\seaborn\\matrix.py:619: ClusterWarning: scipy.cluster: The symmetric non-negative hollow observation matrix looks suspiciously like an uncondensed distance matrix\n",
      "  linkage = hierarchy.linkage(self.array, method=self.method,\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Here we'll average all decks from the same archetype and create a family tree of decks\n",
    "\n",
    "df_mean = df.groupby(['archetype', 'card'], as_index=False).mean()\n",
    "df_mean = df_mean.pivot(index='archetype', columns='card', values='count').fillna(0)\n",
    "df_mean.reset_index()\n",
    "\n",
    "distances = pdist(df_mean.values, metric='braycurtis')\n",
    "DistDf = pd.DataFrame(squareform(distances), columns=df_mean.index, index=df_mean.index)\n",
    "\n",
    "g = sns.clustermap(DistDf, cmap='viridis')\n",
    "plt.setp(g.ax_heatmap.yaxis.get_majorticklabels(), rotation=0)\n",
    "plt.setp(g.ax_heatmap.xaxis.get_majorticklabels(), rotation=90)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 03: Building a classifier\n",
    "\n",
    "We'll first convert our card table into a present-absense matrix, where a one indicates a card is present in a deck (any number of copies) and a zero the card isn't present in a deck.\n"
   ]
  },
  {
   "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>deck</th>\n",
       "      <th>archetype</th>\n",
       "      <th>Abdel Adrian, Gorion's Ward</th>\n",
       "      <th>Abrupt Decay</th>\n",
       "      <th>Abundant Growth</th>\n",
       "      <th>Acererak the Archlich</th>\n",
       "      <th>Ad Nauseam</th>\n",
       "      <th>Aether Channeler</th>\n",
       "      <th>Aether Spellbomb</th>\n",
       "      <th>Aether Vial</th>\n",
       "      <th>...</th>\n",
       "      <th>Windswept Heath</th>\n",
       "      <th>Wirewood Symbiote</th>\n",
       "      <th>Wishclaw Talisman</th>\n",
       "      <th>Witherbloom Apprentice</th>\n",
       "      <th>Wooded Foothills</th>\n",
       "      <th>Wurmcoil Engine</th>\n",
       "      <th>Yavimaya, Cradle of Growth</th>\n",
       "      <th>Young Pyromancer</th>\n",
       "      <th>Zagoth Triome</th>\n",
       "      <th>Ziatora's Proving Ground</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4c Control 1</td>\n",
       "      <td>4c Control</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4c Control 10</td>\n",
       "      <td>4c Control</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>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>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4c Control 11</td>\n",
       "      <td>4c Control</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4c Control 12</td>\n",
       "      <td>4c Control</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>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>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4c Control 13</td>\n",
       "      <td>4c Control</td>\n",
       "      <td>0</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>...</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>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 492 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            deck   archetype  Abdel Adrian, Gorion's Ward  Abrupt Decay  \\\n",
       "0   4c Control 1  4c Control                            0             0   \n",
       "1  4c Control 10  4c Control                            0             0   \n",
       "2  4c Control 11  4c Control                            0             0   \n",
       "3  4c Control 12  4c Control                            0             0   \n",
       "4  4c Control 13  4c Control                            0             1   \n",
       "\n",
       "   Abundant Growth  Acererak the Archlich  Ad Nauseam  Aether Channeler  \\\n",
       "0                1                      0           0                 0   \n",
       "1                0                      0           0                 0   \n",
       "2                1                      0           0                 0   \n",
       "3                0                      0           0                 0   \n",
       "4                0                      0           0                 0   \n",
       "\n",
       "   Aether Spellbomb  Aether Vial  ...  Windswept Heath  Wirewood Symbiote  \\\n",
       "0                 0            0  ...                1                  0   \n",
       "1                 0            0  ...                0                  0   \n",
       "2                 0            0  ...                1                  0   \n",
       "3                 0            0  ...                0                  0   \n",
       "4                 0            0  ...                0                  0   \n",
       "\n",
       "   Wishclaw Talisman  Witherbloom Apprentice  Wooded Foothills  \\\n",
       "0                  0                       0                 1   \n",
       "1                  0                       0                 0   \n",
       "2                  0                       0                 1   \n",
       "3                  0                       0                 0   \n",
       "4                  0                       0                 0   \n",
       "\n",
       "   Wurmcoil Engine  Yavimaya, Cradle of Growth  Young Pyromancer  \\\n",
       "0                0                           0                 0   \n",
       "1                0                           0                 0   \n",
       "2                0                           0                 0   \n",
       "3                0                           0                 0   \n",
       "4                0                           0                 0   \n",
       "\n",
       "   Zagoth Triome  Ziatora's Proving Ground  \n",
       "0              0                         0  \n",
       "1              0                         0  \n",
       "2              0                         0  \n",
       "3              0                         0  \n",
       "4              0                         0  \n",
       "\n",
       "[5 rows x 492 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "card_table_present_absent = card_table.applymap(lambda x: 1 if x > 0 else 0)\n",
    "all_cards = list(card_table_present_absent.columns)\n",
    "card_table_present_absent = pd.merge(archetype_df.reset_index(), card_table_present_absent, left_on='deck', right_on='deck')\n",
    "card_table_present_absent.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we need random subsets of cards from each deck in this format. To do this we will sample a number of random cards from each deck a number of times. This will become our final dataset to train the model on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prepared 1584000 sets for training/testing\n"
     ]
    }
   ],
   "source": [
    "# Build training data\n",
    "\n",
    "deck_data = []\n",
    "\n",
    "for index, row in card_table_present_absent.iterrows():\n",
    "    cards = [k for k,v in row.items() if v == 1]\n",
    "    deck_data.append({\n",
    "        'archetype' : row['archetype'],\n",
    "        'cards': cards\n",
    "    })\n",
    "    \n",
    "\n",
    "\n",
    "from random import shuffle\n",
    "\n",
    "iterations = 1500\n",
    "training_data = []\n",
    "\n",
    "sizes = [4,5,6,7,10,20] # Samples to pick, for small sets of known cards we'll do a few more\n",
    "\n",
    "for card_number in sizes:\n",
    "    for deck in deck_data:\n",
    "        cards = deck['cards']\n",
    "        for i in range(iterations):\n",
    "            shuffle(cards)\n",
    "            selected_cards = cards[:card_number]\n",
    "\n",
    "            training_data.append((deck['archetype'], [1 if c in selected_cards else 0 for c in all_cards]))\n",
    "        \n",
    "shuffle(training_data)\n",
    "\n",
    "print(\"Prepared %d sets for training/testing\" % len(training_data))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we have data suited to train a model for our purpose we'll split the dataset in training and testing data. Next, we'll fit a Random Forest Classifier (RFC) on our training data and check it's performance on the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(max_features=5, n_estimators=40, n_jobs=3)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# build X,y for training, the first n elements, defined by testsize, will be used for testing\n",
    "\n",
    "testsize = 50000\n",
    "\n",
    "X_train = [d[1] for d in training_data[testsize:]]\n",
    "y_train = [d[0] for d in training_data[testsize:]]\n",
    "\n",
    "X_test = [d[1] for d in training_data[:testsize]]\n",
    "y_test = [d[0] for d in training_data[:testsize]]\n",
    "\n",
    "training_data = []\n",
    "\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rfc = RandomForestClassifier(max_depth=None, \n",
    "                             criterion= 'gini', \n",
    "                             max_features= 5, \n",
    "                             n_estimators= 40,\n",
    "                             n_jobs=3)\n",
    "\n",
    "rfc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9794"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfc.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_deck(cards_known):\n",
    "    cards_array = [[1 if c in cards_known else 0 for c in all_cards]]\n",
    "\n",
    "    decks_proba = rfc.predict_proba(cards_array)[0]\n",
    "\n",
    "    decks = pd.DataFrame(list(zip(rfc.classes_, decks_proba)), columns=[\"Deck\", \"Score\"])\\\n",
    "              .sort_values(\"Score\", axis=0, ascending=False)\n",
    "\n",
    "    return decks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's test our classifier on a few samples."
   ]
  },
  {
   "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>Deck</th>\n",
       "      <th>Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Death's Shadow</td>\n",
       "      <td>0.381369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Delver (Other)</td>\n",
       "      <td>0.337440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Aluren</td>\n",
       "      <td>0.256190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Doomsday</td>\n",
       "      <td>0.025000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4c Control</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Other - Aggro</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Jund</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Lands</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Maverick</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Merfolk</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Deck     Score\n",
       "13  Death's Shadow  0.381369\n",
       "14  Delver (Other)  0.337440\n",
       "2           Aluren  0.256190\n",
       "15        Doomsday  0.025000\n",
       "0       4c Control  0.000000\n",
       "29   Other - Aggro  0.000000\n",
       "24            Jund  0.000000\n",
       "25           Lands  0.000000\n",
       "26        Maverick  0.000000\n",
       "27         Merfolk  0.000000"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_deck([\"Polluted Delta\", \"Underground Sea\", \"Brainstorm\", \"Force of Will\"]).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To avoid retraining the data, the classifier is written to disk jusing the gzip and pickle library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gzip, pickle\n",
    "\n",
    "with gzip.open('all_samples.rfc.pickle.gz','wb') as fp:\n",
    "    pickle.dump((all_cards, rfc), fp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 04: Dealing with imbalanced data\n",
    "\n",
    "Some decks only appear once or twice in our dataset, while others are included up to 10 times. This creates an imbalance in our dataset with common decks being overrepresented. This type of imbalance thends to negatively affect a classifiers performance (as it can achieve good classifications always predicting the most common classes). \n",
    "\n",
    "This issue is easily addressed with a packaged **imbalanced learning**. Unfortunately, this isn't part of anaconda and can be installed using\n",
    "\n",
    "```bash\n",
    "conda install -c conda-forge imbalanced-learn\n",
    "```\n",
    "I hope this functionality will be included in sklearn itself as it is an essential component of many machine learning projects."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "\n",
    "rus = RandomUnderSampler()\n",
    "X_rus, y_rus = rus.fit_sample(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By undersampling the abundant classes we do lose a lot of data, this was to be expected. In this case it isn't a big deal as we generated more than enough training data. We'll check if this worked as expected and train a new classifier on the undersampled dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('4c Control', 8687),\n",
       " ('All my Spells', 8687),\n",
       " ('Aluren', 8687),\n",
       " ('Artifacts Blue', 8687),\n",
       " ('BUG Control', 8687),\n",
       " ('BUG Midrange', 8687),\n",
       " ('Bant Control', 8687),\n",
       " ('Burn', 8687),\n",
       " ('Canadian Threshold', 8687),\n",
       " ('Cephalid Breakfast', 8687),\n",
       " ('Cloudpost Ramp', 8687),\n",
       " ('Dark Depths', 8687),\n",
       " ('Death & Taxes', 8687),\n",
       " (\"Death's Shadow\", 8687),\n",
       " ('Delver (Other)', 8687),\n",
       " ('Doomsday', 8687),\n",
       " ('Dragon Stompy', 8687),\n",
       " ('Elves', 8687),\n",
       " ('Enchantress', 8687),\n",
       " ('Goblins', 8687),\n",
       " ('Grixis Control', 8687),\n",
       " ('Hive Mind', 8687),\n",
       " ('Hollow One Madness', 8687),\n",
       " ('Infect', 8687),\n",
       " ('Jund', 8687),\n",
       " ('Lands', 8687),\n",
       " ('Maverick', 8687),\n",
       " ('Merfolk', 8687),\n",
       " ('Mono Black Aggro', 8687),\n",
       " ('Other - Aggro', 8687),\n",
       " ('Other - Combo', 8687),\n",
       " ('Other - Control', 8687),\n",
       " ('Painter', 8687),\n",
       " ('Reanimator', 8687),\n",
       " ('Show and Tell', 8687),\n",
       " ('Stoneblade', 8687),\n",
       " ('Storm', 8687),\n",
       " ('UR Aggro', 8687),\n",
       " ('UWx Control', 8687)]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c = Counter(y_rus)\n",
    "c.most_common()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.95486"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfc = RandomForestClassifier(max_depth=None, \n",
    "                             criterion= 'gini', \n",
    "                             max_features= 5, \n",
    "                             n_estimators= 40,\n",
    "                             n_jobs=3)\n",
    "rfc.fit(X_rus, y_rus)\n",
    "rfc.score(X_test, y_test)"
   ]
  },
  {
   "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>Deck</th>\n",
       "      <th>Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Delver (Other)</td>\n",
       "      <td>0.675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Aluren</td>\n",
       "      <td>0.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Grixis Control</td>\n",
       "      <td>0.075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Death's Shadow</td>\n",
       "      <td>0.050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>Stoneblade</td>\n",
       "      <td>0.050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BUG Control</td>\n",
       "      <td>0.025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Doomsday</td>\n",
       "      <td>0.025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4c Control</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Lands</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Maverick</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Deck  Score\n",
       "14  Delver (Other)  0.675\n",
       "2           Aluren  0.100\n",
       "20  Grixis Control  0.075\n",
       "13  Death's Shadow  0.050\n",
       "35      Stoneblade  0.050\n",
       "4      BUG Control  0.025\n",
       "15        Doomsday  0.025\n",
       "0       4c Control  0.000\n",
       "25           Lands  0.000\n",
       "26        Maverick  0.000"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_deck([\"Polluted Delta\", \"Underground Sea\", \"Brainstorm\", \"Force of Will\"]).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gzip, pickle\n",
    "\n",
    "with gzip.open('balanced_samples.rfc.pickle.gz','wb') as fp:\n",
    "    pickle.dump((all_cards, rfc), fp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Step 05: Load classifier and play with predictions\n",
    "\n",
    "\n",
    "In this final part the classifier will be loaded from a pickle and you can play with predictions.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gzip, pickle\n",
    "import pandas as pd\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "\n",
    "with gzip.open('balanced_samples.rfc.pickle.gz','rb') as fp:\n",
    "    (all_cards, rfc) = pickle.load(fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_deck(cards_known):\n",
    "    \"\"\"\n",
    "    This will take a list of known cards and convert it in a matrix compatible with the classifier.\n",
    "    Next, this classifier will be used to predict which deck the known cards are coming from and \n",
    "    return the results as a sorted pandas dataframe.\n",
    "    \"\"\"\n",
    "    cards_array = [[1 if c in cards_known else 0 for c in all_cards]]\n",
    "\n",
    "    cards_not_in_model = [c for c in cards_known if c not in all_cards]\n",
    "    \n",
    "    if len(cards_not_in_model) > 0:\n",
    "        print(\"Some cards were not included when trainig the model, these will be ignored: %s\" % ','.join(cards_not_in_model))\n",
    "    \n",
    "    decks_proba = rfc.predict_proba(cards_array)[0]\n",
    "\n",
    "    decks = pd.DataFrame(list(zip(rfc.classes_, decks_proba)), columns=[\"Deck\", \"Score\"])\\\n",
    "              .sort_values(\"Score\", axis=0, ascending=False)\n",
    "\n",
    "    return decks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck([\"Plains\", \"Mother of Runes\", \"Wasteland\"]).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck([\"Dark Confidant\", \"Forest\", \"Mox Diamond\", \"Swamp\"]).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck([\"Arcum's Astrolabe\", \"Snow-Covered Forest\", \"Brainstorm\", \"Noble Hierarch\"]).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck([\"Polluted Delta\", \"Volcanic Island\", \"Underground Sea\", \"Brainstorm\"]).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck(['Birds of Paradise', \"Snow-Covered Forest\", \"Windswept Heath\", \"Misthollow Griffin\", \"Force of Will\"]).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck([\"Arcum's Astrolabe\", \"Snow-Covered Swamp\", \"Brainstorm\", \"Polluted Delta\", \"Snow-Covered Island\", \"Ponder\"]).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_deck([\"Arcum's Astrolabe\", \"Snow-Covered Swamp\", \"Polluted Delta\",  \"Snow-Covered Forest\", \"Brainstorm\", \"Ponder\"]).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "This seems to work very well and can easily be adapted for other formats as well. Do note that manual curation could improve these prediction. For instance the UWx Control decks are very diverse and the classifier would likely perform better if this class was removed.\n",
    "\n",
    "Overall a cool application of machine learning I haven't seen before ! \n"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}