ch03.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn import (\n",
    "    ensemble,\n",
    "    model_selection,    \n",
    "    preprocessing,\n",
    "    tree,\n",
    ")\n",
    "from sklearn.metrics import (\n",
    "    auc,\n",
    "    confusion_matrix,\n",
    "    roc_auc_score,\n",
    "    roc_curve,\n",
    ")\n",
    "from sklearn.model_selection import (\n",
    "    train_test_split,\n",
    "    StratifiedKFold,\n",
    ")\n",
    "from yellowbrick.classifier import (\n",
    "    ConfusionMatrix,\n",
    "    ROCAUC,\n",
    ")\n",
    "from yellowbrick.model_selection import (\n",
    "    LearningCurve,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "url = (\n",
    "    \"file:\" \n",
    "    \"titanic3.xls\"\n",
    ")\n",
    "df = pd.read_excel(url)\n",
    "orig_df = df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass         int64\n",
       "survived       int64\n",
       "name          object\n",
       "sex           object\n",
       "age          float64\n",
       "sibsp          int64\n",
       "parch          int64\n",
       "ticket        object\n",
       "fare         float64\n",
       "cabin         object\n",
       "embarked      object\n",
       "boat          object\n",
       "body         float64\n",
       "home.dest     object\n",
       "dtype: object"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import pandas_profiling\n",
    "#pandas_profiling.ProfileReport(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1309, 14)"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "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>pclass</th>\n",
       "      <th>survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1309.000000</td>\n",
       "      <td>1309.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2.294882</td>\n",
       "      <td>0.381971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.837836</td>\n",
       "      <td>0.486055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            pclass     survived\n",
       "count  1309.000000  1309.000000\n",
       "mean      2.294882     0.381971\n",
       "std       0.837836     0.486055\n",
       "min       1.000000     0.000000\n",
       "25%       2.000000     0.000000\n",
       "50%       3.000000     0.000000\n",
       "75%       3.000000     1.000000\n",
       "max       3.000000     1.000000"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe().iloc[:, :2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass          0\n",
       "survived        0\n",
       "name            0\n",
       "sex             0\n",
       "age           263\n",
       "sibsp           0\n",
       "parch           0\n",
       "ticket          0\n",
       "fare            1\n",
       "cabin        1014\n",
       "embarked        2\n",
       "boat          823\n",
       "body         1188\n",
       "home.dest     564\n",
       "dtype: int64"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     1\n",
       "1     1\n",
       "2     2\n",
       "3     1\n",
       "4     2\n",
       "5     1\n",
       "6     1\n",
       "7     2\n",
       "8     1\n",
       "9     2\n",
       "10    1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum(axis=1).loc[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "mask = df.isnull().any(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    True\n",
       "1    True\n",
       "2    True\n",
       "3    True\n",
       "4    True\n",
       "dtype: bool"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask.head()  # rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      NaN\n",
       "1      NaN\n",
       "2      NaN\n",
       "3    135.0\n",
       "4      NaN\n",
       "Name: body, dtype: float64"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[mask].body.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sex\n",
       "male      843\n",
       "female    466\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sex.value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "embarked\n",
       "S      914\n",
       "C      270\n",
       "Q      123\n",
       "NaN      2\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.embarked.value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     Allen, Miss. Elisabeth Walton\n",
       "1    Allison, Master. Hudson Trevor\n",
       "2      Allison, Miss. Helen Loraine\n",
       "Name: name, dtype: object"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name = df.name\n",
    "name.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(\n",
    "    columns=[\n",
    "        \"name\",\n",
    "        \"ticket\",\n",
    "        \"home.dest\",\n",
    "        \"boat\",\n",
    "        \"body\",\n",
    "        \"cabin\",\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.get_dummies(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['pclass', 'survived', 'age', 'sibsp', 'parch', 'fare', 'sex_female',\n",
       "       'sex_male', 'embarked_C', 'embarked_Q', 'embarked_S'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(columns=\"sex_male\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.get_dummies(df, drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['pclass', 'survived', 'age', 'sibsp', 'parch', 'fare', 'sex_female',\n",
       "       'embarked_C', 'embarked_Q', 'embarked_S'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df.survived\n",
    "X = df.drop(columns=\"survived\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import janitor as jn\n",
    "#X, y = jn.get_features_targets(\n",
    "#    df, target_columns=\"survived\"\n",
    "#)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = model_selection.train_test_split(\n",
    "    X, y, test_size=0.3, random_state=42\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['pclass', 'age', 'sibsp', 'parch', 'fare', 'sex_female', 'embarked_C',\n",
       "       'embarked_Q', 'embarked_S'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.experimental import (\n",
    "    enable_iterative_imputer,\n",
    ")\n",
    "from sklearn import impute\n",
    "num_cols = [\n",
    "    \"pclass\",\n",
    "    \"age\",\n",
    "    \"sibsp\",\n",
    "    \"parch\",\n",
    "    \"fare\",\n",
    "    \"sex_female\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "imputer = impute.IterativeImputer()\n",
    "imputed = imputer.fit_transform(\n",
    "    X_train[num_cols]\n",
    ")\n",
    "X_train.loc[:, num_cols] = imputed\n",
    "imputed = imputer.transform(X_test[num_cols])\n",
    "X_test.loc[:, num_cols] = imputed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [],
   "source": [
    "meds = X_train.median()\n",
    "X_train = X_train.fillna(meds)\n",
    "X_test = X_test.fillna(meds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {
    "lines_to_next_cell": 1
   },
   "outputs": [],
   "source": [
    "cols = ['pclass', 'age', 'sibsp', 'parch', 'fare', 'sex_female', 'embarked_C',\n",
    "   'embarked_Q', 'embarked_S']\n",
    "sca = preprocessing.StandardScaler()\n",
    "X_train = sca.fit_transform(X_train)\n",
    "X_train = pd.DataFrame(X_train, columns=cols)\n",
    "X_test = sca.transform(X_test)\n",
    "X_test = pd.DataFrame(X_test, columns=cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "lines_to_next_cell": 1
   },
   "outputs": [],
   "source": [
    "def tweak_titanic(df):\n",
    "    df = df.drop(\n",
    "        columns=[\n",
    "            \"name\",\n",
    "            \"ticket\",\n",
    "            \"home.dest\",\n",
    "            \"boat\",\n",
    "            \"body\",\n",
    "            \"cabin\",\n",
    "        ]\n",
    "    ).pipe(pd.get_dummies, drop_first=True)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_train_test_X_y(\n",
    "    df, y_col, size=0.3, std_cols=None\n",
    "):\n",
    "    y = df[y_col]\n",
    "    X = df.drop(columns=y_col)\n",
    "    X_train, X_test, y_train, y_test = model_selection.train_test_split(\n",
    "        X, y, test_size=size, random_state=42\n",
    "    )\n",
    "    cols = X.columns\n",
    "    num_cols = [\n",
    "        \"pclass\",\n",
    "        \"age\",\n",
    "        \"sibsp\",\n",
    "        \"parch\",\n",
    "        \"fare\",\n",
    "    ]\n",
    "    fi = impute.IterativeImputer()\n",
    "    fitted = fi.fit_transform(X_train[num_cols])\n",
    "    X_train = X_train.assign(**{c:fitted[:,i] for i, c in enumerate(num_cols)})\n",
    "    test_fit = fi.transform(X_test[num_cols])\n",
    "    X_test = X_test.assign(**{c:test_fit[:,i] for i, c in enumerate(num_cols)})\n",
    "    if std_cols:\n",
    "        std = preprocessing.StandardScaler()\n",
    "        fitted = std.fit_transform(X_train[std_cols])\n",
    "        X_train = X_train.assign(**{c:fitted[:,i] for i, c in enumerate(std_cols)})\n",
    "        test_fit = std.transform(X_test[std_cols])\n",
    "        X_test = X_test.assign(**{c:test_fit[:,i] for i, c in enumerate(std_cols)})\n",
    "\n",
    "    return X_train, X_test, y_train, y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "ti_df = tweak_titanic(orig_df)\n",
    "std_cols = \"pclass,age,sibsp,fare\".split(\",\")\n",
    "X_train, X_test, y_train, y_test = get_train_test_X_y(\n",
    "    ti_df, \"survived\", std_cols=std_cols\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5699745547073791"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.dummy import DummyClassifier\n",
    "bm = DummyClassifier()\n",
    "bm.fit(X_train, y_train)\n",
    "bm.score(X_test, y_test)  # accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_classification.py:1471: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 due to no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 166,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn import metrics\n",
    "metrics.precision_score(\n",
    "    y_test, bm.predict(X_test)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = pd.concat([X_train, X_test])\n",
    "y = pd.concat([y_train, y_test])\n",
    "from sklearn import model_selection\n",
    "from sklearn.dummy import DummyClassifier\n",
    "from sklearn.linear_model import (\n",
    "    LogisticRegression,\n",
    ")\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import (\n",
    "    KNeighborsClassifier,\n",
    ")\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import (\n",
    "    RandomForestClassifier,\n",
    ")\n",
    "import xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DummyClassifier         AUC: 0.500 STD: 0.00\n",
      "LogisticRegression      AUC: 0.843 STD: 0.03\n",
      "DecisionTreeClassifier  AUC: 0.769 STD: 0.03\n",
      "KNeighborsClassifier    AUC: 0.828 STD: 0.05\n",
      "GaussianNB              AUC: 0.817 STD: 0.04\n",
      "SVC                     AUC: 0.837 STD: 0.05\n",
      "RandomForestClassifier  AUC: 0.849 STD: 0.03\n",
      "XGBClassifier           AUC: 0.857 STD: 0.03\n"
     ]
    }
   ],
   "source": [
    "for model in [\n",
    "    DummyClassifier,\n",
    "    LogisticRegression,\n",
    "    DecisionTreeClassifier,\n",
    "    KNeighborsClassifier,\n",
    "    GaussianNB,\n",
    "    SVC,\n",
    "    RandomForestClassifier,\n",
    "    xgboost.XGBClassifier,\n",
    "]:\n",
    "    cls = model()\n",
    "    kfold = model_selection.KFold(\n",
    "        n_splits=10, random_state=None\n",
    "    )\n",
    "    s = model_selection.cross_val_score(\n",
    "        cls, X, y, scoring=\"roc_auc\", cv=kfold\n",
    "    )\n",
    "    print(\n",
    "        f\"{model.__name__:22}  AUC: \"\n",
    "        f\"{s.mean():.3f} STD: {s.std():.2f}\"\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "StackingClassifier  AUC: nan  STD: nan\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:839: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 459, in _score\n",
      "    y_pred = method_caller(clf, \"decision_function\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "During handling of the above exception, another exception occurred:\n",
      "\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 136, in __call__\n",
      "    score = scorer._score(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 466, in _score\n",
      "    y_pred = method_caller(clf, \"predict_proba\", X, pos_label=pos_label)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/metrics/_scorer.py\", line 86, in _cached_call\n",
      "    result, _ = _get_response_values(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_response.py\", line 182, in _get_response_values\n",
      "    classes = estimator.classes_\n",
      "AttributeError: 'StackingClassifier' object has no attribute 'classes_'\n",
      "\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "from mlxtend.classifier import (\n",
    "    StackingClassifier,\n",
    ")\n",
    "clfs = [\n",
    "    x()\n",
    "    for x in [\n",
    "        LogisticRegression,\n",
    "        DecisionTreeClassifier,\n",
    "        KNeighborsClassifier,\n",
    "        GaussianNB,\n",
    "        SVC,\n",
    "        RandomForestClassifier,\n",
    "    ]\n",
    "]\n",
    "stack = StackingClassifier(\n",
    "    classifiers=clfs,\n",
    "    meta_classifier=LogisticRegression(),\n",
    ")\n",
    "kfold = model_selection.KFold(\n",
    "    n_splits=10, random_state=None\n",
    ")\n",
    "s = model_selection.cross_val_score(\n",
    "    stack, X, y, scoring=\"roc_auc\", cv=kfold\n",
    ")\n",
    "print(\n",
    "    f\"{stack.__class__.__name__}  \"\n",
    "    f\"AUC: {s.mean():.3f}  STD: {s.std():.2f}\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>RandomForestClassifier(random_state=42)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestClassifier</label><div class=\"sk-toggleable__content\"><pre>RandomForestClassifier(random_state=42)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "RandomForestClassifier(random_state=42)"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf = ensemble.RandomForestClassifier(\n",
    "    n_estimators=100, random_state=42\n",
    ")\n",
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7837150127226463"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7916666666666666"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.precision_score(\n",
    "    y_test, rf.predict(X_test)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age            0.285\n",
      "fare           0.262\n",
      "sex_male       0.241\n",
      "pclass         0.089\n",
      "sibsp          0.050\n"
     ]
    }
   ],
   "source": [
    "for col, val in sorted(\n",
    "    zip(\n",
    "        X_train.columns,\n",
    "        rf.feature_importances_,\n",
    "    ),\n",
    "    key=lambda x: x[1],\n",
    "    reverse=True,\n",
    ")[:5]:\n",
    "    print(f\"{col:10}{val:10.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py:425: FitFailedWarning: \n",
      "20 fits failed out of a total of 40.\n",
      "The score on these train-test partitions for these parameters will be set to nan.\n",
      "If these failures are not expected, you can try to debug them by setting error_score='raise'.\n",
      "\n",
      "Below are more details about the failures:\n",
      "--------------------------------------------------------------------------------\n",
      "10 fits failed with the following error:\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py\", line 729, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/base.py\", line 1145, in wrapper\n",
      "    estimator._validate_params()\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/base.py\", line 638, in _validate_params\n",
      "    validate_parameter_constraints(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_param_validation.py\", line 96, in validate_parameter_constraints\n",
      "    raise InvalidParameterError(\n",
      "sklearn.utils._param_validation.InvalidParameterError: The 'max_features' parameter of RandomForestClassifier must be an int in the range [1, inf), a float in the range (0.0, 1.0], a str among {'sqrt', 'log2'} or None. Got 'auto' instead.\n",
      "\n",
      "--------------------------------------------------------------------------------\n",
      "10 fits failed with the following error:\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_validation.py\", line 729, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/base.py\", line 1145, in wrapper\n",
      "    estimator._validate_params()\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/base.py\", line 638, in _validate_params\n",
      "    validate_parameter_constraints(\n",
      "  File \"/home/codespace/.local/lib/python3.10/site-packages/sklearn/utils/_param_validation.py\", line 96, in validate_parameter_constraints\n",
      "    raise InvalidParameterError(\n",
      "sklearn.utils._param_validation.InvalidParameterError: The 'max_features' parameter of RandomForestClassifier must be an int in the range [1, inf), a float in the range (0.0, 1.0], a str among {'log2', 'sqrt'} or None. Got 'auto' instead.\n",
      "\n",
      "  warnings.warn(some_fits_failed_message, FitFailedWarning)\n",
      "/home/codespace/.local/lib/python3.10/site-packages/sklearn/model_selection/_search.py:979: UserWarning: One or more of the test scores are non-finite: [0.7849608  0.7893205  0.76312663 0.77077691        nan        nan\n",
      "        nan        nan]\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'max_features': 0.4, 'min_samples_leaf': 1, 'n_estimators': 200, 'random_state': 42}\n"
     ]
    }
   ],
   "source": [
    "rf4 = ensemble.RandomForestClassifier()\n",
    "params = {\n",
    "    \"max_features\": [0.4, \"auto\"],\n",
    "    \"n_estimators\": [15, 200],\n",
    "    \"min_samples_leaf\": [1, 0.1],\n",
    "    \"random_state\": [42],\n",
    "}\n",
    "cv = model_selection.GridSearchCV(\n",
    "    rf4, params, n_jobs=-1\n",
    ").fit(X_train, y_train)\n",
    "print(cv.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7531806615776081"
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf5 = ensemble.RandomForestClassifier(\n",
    "    **{\n",
    "        \"max_features\": None,\n",
    "        \"min_samples_leaf\": 0.1,\n",
    "        \"n_estimators\": 200,\n",
    "        \"random_state\": 42,\n",
    "    }\n",
    ")\n",
    "rf5.fit(X_train, y_train)\n",
    "rf5.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[218,   6],\n",
       "       [ 91,  78]])"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "y_pred = rf5.predict(X_test)\n",
    "confusion_matrix(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/matt/.env/364/lib/python3.6/site-packages/sklearn/metrics/classification.py:182: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  score = y_true == y_pred\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mapping = {0: \"died\", 1: \"survived\"}\n",
    "fig, ax = plt.subplots(figsize=(6, 6))\n",
    "cm_viz = ConfusionMatrix(\n",
    "    rf5,\n",
    "    classes=[\"died\", \"survived\"],\n",
    "    label_encoder=mapping,\n",
    ")\n",
    "cm_viz.score(X_test, y_test)\n",
    "cm_viz.poof()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7173763736263736"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = rf5.predict(X_test)\n",
    "roc_auc_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(6, 6))\n",
    "roc_viz = ROCAUC(rf5)\n",
    "roc_viz.score(X_test, y_test)\n",
    "roc_viz.poof()\n",
    "#fig.savefig(\"images/mlpr_0305.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "fig, ax = plt.subplots(figsize=(6, 4))\n",
    "cv = StratifiedKFold(12)\n",
    "sizes = np.linspace(0.3, 1.0, 10)\n",
    "lc_viz = LearningCurve(\n",
    "    rf5,\n",
    "    cv=cv,\n",
    "    train_sizes=sizes,\n",
    "    scoring=\"f1_weighted\",\n",
    "    n_jobs=4,\n",
    "    ax=ax,\n",
    ")\n",
    "lc_viz.fit(X, y)\n",
    "lc_viz.poof()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {
    "lines_to_next_cell": 2
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7173763736263736"
      ]
     },
     "execution_count": 181,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pickle\n",
    "pic = pickle.dumps(rf5)\n",
    "rf6 = pickle.loads(pic)\n",
    "y_pred = rf6.predict(X_test)\n",
    "roc_auc_score(y_test, y_pred)"
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "formats": "ipynb,py:light"
  },
  "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}