{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SAMPO/FAB"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## データ準備"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "まず、学習用のデータを確認します。\n",
    "\n",
    "本ノートブックでは、中古車の販売価格を示すデータを使用します。\n",
    "\n",
    "データの詳細については、以下をご覧ください。\n",
    "\n",
    "[Automobile Data Set][1]\n",
    "\n",
    "[1]:https://archive.ics.uci.edu/ml/datasets/Automobile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "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>_sid</th>\n",
       "      <th>price</th>\n",
       "      <th>symboling</th>\n",
       "      <th>normalized-losses</th>\n",
       "      <th>num-of-doors</th>\n",
       "      <th>wheel-base</th>\n",
       "      <th>length</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>12964</td>\n",
       "      <td>3</td>\n",
       "      <td>145.0</td>\n",
       "      <td>two</td>\n",
       "      <td>95.9</td>\n",
       "      <td>173.2</td>\n",
       "      <td>66.3</td>\n",
       "      <td>50.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>10198</td>\n",
       "      <td>0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>four</td>\n",
       "      <td>97.0</td>\n",
       "      <td>173.5</td>\n",
       "      <td>65.4</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>11245</td>\n",
       "      <td>0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>four</td>\n",
       "      <td>98.8</td>\n",
       "      <td>177.8</td>\n",
       "      <td>66.5</td>\n",
       "      <td>55.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>14399</td>\n",
       "      <td>0</td>\n",
       "      <td>108.0</td>\n",
       "      <td>four</td>\n",
       "      <td>100.4</td>\n",
       "      <td>184.6</td>\n",
       "      <td>66.5</td>\n",
       "      <td>56.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>25552</td>\n",
       "      <td>-1</td>\n",
       "      <td>93.0</td>\n",
       "      <td>four</td>\n",
       "      <td>110.0</td>\n",
       "      <td>190.9</td>\n",
       "      <td>70.3</td>\n",
       "      <td>56.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   _sid  price  symboling  normalized-losses num-of-doors  wheel-base  length  \\\n",
       "0     0  12964          3              145.0          two        95.9   173.2   \n",
       "1     1  10198          0               89.0         four        97.0   173.5   \n",
       "2     2  11245          0              115.0         four        98.8   177.8   \n",
       "3     3  14399          0              108.0         four       100.4   184.6   \n",
       "4     4  25552         -1               93.0         four       110.0   190.9   \n",
       "\n",
       "   width  height  \n",
       "0   66.3    50.2  \n",
       "1   65.4    53.0  \n",
       "2   66.5    55.5  \n",
       "3   66.5    56.1  \n",
       "4   70.3    56.5  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "train_data = pd.read_csv('../data/train_data.csv')\n",
    "train_data.insert(0, '_sid', list(range(train_data.shape[0])))\n",
    "train_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "上記のデータについて、以下の前処理を実施します。\n",
    "\n",
    "- num-of-doorsはカテゴリ変数のため、二値展開\n",
    "- num-of-doorsを除く変数は数値データであるため、標準化\n",
    "\n",
    "これらの前処理を、後ほどSPDで定義します。"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "次に、ASDを作成します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>scale</th>\n",
       "      <th>domain</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>_sid</th>\n",
       "      <td>INTEGER</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price</th>\n",
       "      <td>INTEGER</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>symboling</th>\n",
       "      <td>INTEGER</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>normalized-losses</th>\n",
       "      <td>REAL</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>num-of-doors</th>\n",
       "      <td>NOMINAL</td>\n",
       "      <td>[four, two]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wheel-base</th>\n",
       "      <td>REAL</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>length</th>\n",
       "      <td>REAL</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>width</th>\n",
       "      <td>REAL</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>height</th>\n",
       "      <td>REAL</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     scale       domain\n",
       "_sid               INTEGER          NaN\n",
       "price              INTEGER          NaN\n",
       "symboling          INTEGER          NaN\n",
       "normalized-losses     REAL          NaN\n",
       "num-of-doors       NOMINAL  [four, two]\n",
       "wheel-base            REAL          NaN\n",
       "length                REAL          NaN\n",
       "width                 REAL          NaN\n",
       "height                REAL          NaN"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sampotools.api import gen_asd_from_pandas_df\n",
    "\n",
    "asd = gen_asd_from_pandas_df(train_data)\n",
    "pd.DataFrame(asd).T[['scale', 'domain']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "最後に、予測用データを読み込みます。"
   ]
  },
  {
   "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>_sid</th>\n",
       "      <th>price</th>\n",
       "      <th>symboling</th>\n",
       "      <th>normalized-losses</th>\n",
       "      <th>num-of-doors</th>\n",
       "      <th>wheel-base</th>\n",
       "      <th>length</th>\n",
       "      <th>width</th>\n",
       "      <th>height</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>6338</td>\n",
       "      <td>1</td>\n",
       "      <td>87.0</td>\n",
       "      <td>two</td>\n",
       "      <td>95.7</td>\n",
       "      <td>158.7</td>\n",
       "      <td>63.6</td>\n",
       "      <td>54.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>18150</td>\n",
       "      <td>3</td>\n",
       "      <td>150.0</td>\n",
       "      <td>two</td>\n",
       "      <td>99.1</td>\n",
       "      <td>186.6</td>\n",
       "      <td>66.5</td>\n",
       "      <td>56.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>17669</td>\n",
       "      <td>2</td>\n",
       "      <td>134.0</td>\n",
       "      <td>two</td>\n",
       "      <td>98.4</td>\n",
       "      <td>176.2</td>\n",
       "      <td>65.6</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>11549</td>\n",
       "      <td>2</td>\n",
       "      <td>134.0</td>\n",
       "      <td>two</td>\n",
       "      <td>98.4</td>\n",
       "      <td>176.2</td>\n",
       "      <td>65.6</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>32250</td>\n",
       "      <td>0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>four</td>\n",
       "      <td>113.0</td>\n",
       "      <td>199.6</td>\n",
       "      <td>69.6</td>\n",
       "      <td>52.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   _sid  price  symboling  normalized-losses num-of-doors  wheel-base  length  \\\n",
       "0     0   6338          1               87.0          two        95.7   158.7   \n",
       "1     1  18150          3              150.0          two        99.1   186.6   \n",
       "2     2  17669          2              134.0          two        98.4   176.2   \n",
       "3     3  11549          2              134.0          two        98.4   176.2   \n",
       "4     4  32250          0              145.0         four       113.0   199.6   \n",
       "\n",
       "   width  height  \n",
       "0   63.6    54.5  \n",
       "1   66.5    56.1  \n",
       "2   65.6    53.0  \n",
       "3   65.6    52.0  \n",
       "4   69.6    52.8  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_data = pd.read_csv('../data/predict_data.csv')\n",
    "predict_data.insert(0, '_sid', list(range(predict_data.shape[0])))\n",
    "predict_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 分析手順の設計"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "目的変数priceは連続値であるため、回帰用のcomponentである「FABHMEBernGateLinearRg Component」を選択します。\n",
    "\n",
    "学習を行う際に、以下のパラメーターを指定します。\n",
    "\n",
    "seed\n",
    "\n",
    "- componentで使用される乱数シードです。\n",
    "\n",
    "tree_depth\n",
    "\n",
    "- 門木の深さを設定するパラメーターです。\n",
    "\n",
    "shrink_threshold\n",
    "\n",
    "- 門木の枝刈りの閾値を設定するパラメーターです。\n",
    "\n",
    "seed, tree_depth, shrink_thresholdの詳細は、「SAMPO Reference V1.6.2」をご覧ください。<br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SPD定義"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sampo.api import gen_spd\n",
    "\n",
    "# seedの指定\n",
    "seed = 0\n",
    "\n",
    "# SPD の定義\n",
    "spd_content = '''\n",
    "dl -> std -> rg\n",
    "   -> bexp -> rg\n",
    "\n",
    "---\n",
    "components:\n",
    "    dl:\n",
    "        component: DataLoader\n",
    "\n",
    "    std:\n",
    "        component: StandardizeFDComponent\n",
    "        features: scale == 'real' or scale == 'integer'\n",
    "\n",
    "    bexp:\n",
    "        component: BinaryExpandFDComponent\n",
    "        features: scale == 'nominal'\n",
    "\n",
    "    rg:\n",
    "        component: FABHMEBernGateLinearRgComponent\n",
    "        features: name != 'price'\n",
    "        target: name == 'price'\n",
    "        standardize_target: True\n",
    "        tree_depth: 3\n",
    "        shrink_threshold: 2.0\n",
    "\n",
    "global_settings:\n",
    "    keep_attributes:\n",
    "        - price\n",
    "    feature_exclude:\n",
    "        - price\n",
    "'''\n",
    "\n",
    "spd = gen_spd(template=spd_content)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SRC定義"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sampo.api import gen_src\n",
    "\n",
    "train_src_temp = '''\n",
    "train:\n",
    "    type: learn\n",
    "    data_sources:\n",
    "        dl:\n",
    "            df: {{ data_df }}\n",
    "            attr_schema: {{ asd }}\n",
    "'''\n",
    "\n",
    "predict_src_temp = '''\n",
    "predict:\n",
    "    type: predict\n",
    "    data_sources:\n",
    "        dl:\n",
    "            df: {{ data_df }}\n",
    "            attr_schema: {{ asd }}\n",
    "    model_process: {{ model_process }}\n",
    "'''"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 学習、予測"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "学習を実行します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "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>process name</th>\n",
       "      <th>version</th>\n",
       "      <th>started at</th>\n",
       "      <th>running time</th>\n",
       "      <th>status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>train</td>\n",
       "      <td>4a53449a-801c-415f-a9e4-a2e9b47c4979</td>\n",
       "      <td>2020-03-05 13:53:44.265040</td>\n",
       "      <td>00:00:01.261934</td>\n",
       "      <td>Succeeded</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  process name                               version  \\\n",
       "0        train  4a53449a-801c-415f-a9e4-a2e9b47c4979   \n",
       "\n",
       "                  started at    running time     status  \n",
       "0 2020-03-05 13:53:44.265040 00:00:01.261934  Succeeded  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sampo.api import process_store, process_runner\n",
    "\n",
    "train_src = gen_src(template=train_src_temp, params={'data_df': train_data, 'asd': asd})\n",
    "\n",
    "pstore_url = 'pstore_rg'\n",
    "!rm -rf $pstore_url\n",
    "process_store.create(pstore_url)\n",
    "\n",
    "process_runner.run(spd=spd, src=train_src, pstore_url=pstore_url, seed=seed)\n",
    "process_store.list_process_metadata(pstore_url)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "予測を実行します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "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>process name</th>\n",
       "      <th>version</th>\n",
       "      <th>started at</th>\n",
       "      <th>running time</th>\n",
       "      <th>status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>predict</td>\n",
       "      <td>b11be9f7-5ef6-4cd9-994d-6f661864f931</td>\n",
       "      <td>2020-03-05 13:53:45.625547</td>\n",
       "      <td>00:00:00.736196</td>\n",
       "      <td>Succeeded</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>train</td>\n",
       "      <td>4a53449a-801c-415f-a9e4-a2e9b47c4979</td>\n",
       "      <td>2020-03-05 13:53:44.265040</td>\n",
       "      <td>00:00:01.261934</td>\n",
       "      <td>Succeeded</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  process name                               version  \\\n",
       "0      predict  b11be9f7-5ef6-4cd9-994d-6f661864f931   \n",
       "1        train  4a53449a-801c-415f-a9e4-a2e9b47c4979   \n",
       "\n",
       "                  started at    running time     status  \n",
       "0 2020-03-05 13:53:45.625547 00:00:00.736196  Succeeded  \n",
       "1 2020-03-05 13:53:44.265040 00:00:01.261934  Succeeded  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_proc_name = 'train'\n",
    "predict_src = gen_src(template=predict_src_temp, params={'model_process': train_proc_name,\n",
    "                                                         'data_df': predict_data, 'asd': asd})\n",
    "process_runner.run(src=predict_src, pstore_url=pstore_url)\n",
    "process_store.list_process_metadata(pstore_url)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## モデルバリデーション"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 予測精度の確認"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "作成したモデルを評価するため、予測精度を確認します。\n",
    "\n",
    "評価指標として、RMSEを選定します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7621801464683062"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predict_proc_name = 'predict'\n",
    "with process_store.open_process(pstore_url, predict_proc_name) as prl:\n",
    "    prl = prl\n",
    "prl.load_comp_output_evaluation('rg')['std_root_mean_squared_error'][0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 門木の可視化"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "モデルがもつ門木を可視化します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.display import display, Image\n",
    "from sampovis.api.fabhme_vis import save_gate_tree\n",
    "\n",
    "save_gate_tree(pstore_url, train_proc_name, 'output')\n",
    "display(Image('output/{}_rg_fabhme_gate_tree.png'.format(train_proc_name)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 予測式の可視化"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "モデルがもつ予測式情報にアクセスし、可視化します。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "model_params = prl.load_model('rg')\n",
    "prediction_formulas = model_params['prediction_formulas']\n",
    "relevant_feature_indices = prediction_formulas.sum(axis=1) != 0\n",
    "prediction_formulas = prediction_formulas[relevant_feature_indices]\n",
    "prediction_formulas.plot(kind='barh', figsize=(8, 4), stacked=True)\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "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": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}
