{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "9b10c236",
   "metadata": {},
   "source": [
    "# Shapash in Jupyter - Imbalanced Titanic Classification\n",
    "\n",
    "In this tutorial you will:\n",
    "- Build a binary classifier on an imbalanced Titanic dataset\n",
    "- Compare a baseline model and a weighted model\n",
    "- Focus on precision, recall, and F1-score for the minority class\n",
    "- Explain the weighted model with Shapash and prepare the webapp launch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "91e746de",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from category_encoders import one_hot\n",
    "from lightgbm import LGBMClassifier\n",
    "from sklearn.metrics import (\n",
    "    classification_report,\n",
    "    precision_score,\n",
    "    recall_score,\n",
    "    f1_score,\n",
    "    accuracy_score,\n",
    ")\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from shapash import SmartExplainer\n",
    "from shapash.data.data_loader import data_loading"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2270005a",
   "metadata": {},
   "source": [
    "## 1. Load and prepare Titanic data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4b2cd45e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "PassengerId",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "Pclass",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "Age",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "Sex",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "SibSp",
         "rawType": "int64",
         "type": "integer"
        },
        {
         "name": "Parch",
         "rawType": "int64",
         "type": "integer"
        }
       ],
       "ref": "04f82503-4c03-4a3e-a269-bff5f68c9cc4",
       "rows": [
        [
         "1",
         "3",
         "22.0",
         "male",
         "1",
         "0"
        ],
        [
         "2",
         "1",
         "38.0",
         "female",
         "1",
         "0"
        ],
        [
         "3",
         "3",
         "26.0",
         "female",
         "0",
         "0"
        ],
        [
         "4",
         "1",
         "35.0",
         "female",
         "1",
         "0"
        ],
        [
         "5",
         "3",
         "35.0",
         "male",
         "0",
         "0"
        ]
       ],
       "shape": {
        "columns": 5,
        "rows": 5
       }
      },
      "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>Age</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>male</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>female</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>male</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pclass   Age     Sex  SibSp  Parch\n",
       "PassengerId                                    \n",
       "1                 3  22.0    male      1      0\n",
       "2                 1  38.0  female      1      0\n",
       "3                 3  26.0  female      0      0\n",
       "4                 1  35.0  female      1      0\n",
       "5                 3  35.0    male      0      0"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_df, titanic_dict = data_loading(\"titanic\")\n",
    "\n",
    "# Same feature engineering style as webapp_launch.py\n",
    "titanic_df[\"Pclass\"] = titanic_df[\"Pclass\"].map({\"First class\": 1, \"Second class\": 2, \"Third class\": 3})\n",
    "\n",
    "features = [\"Pclass\", \"Age\", \"Sex\", \"SibSp\", \"Parch\"]\n",
    "target_name = \"Survived\"\n",
    "\n",
    "X = titanic_df[features]\n",
    "y = titanic_df[target_name].to_frame()\n",
    "\n",
    "feature_dict = {\n",
    "    \"Pclass\": \"Ticket class\",\n",
    "    \"Age\": \"Age\",\n",
    "    \"Sex\": \"Sex\",\n",
    "    \"SibSp\": \"Number of siblings/spouses aboard\",\n",
    "    \"Parch\": \"Number of parents/children aboard\",\n",
    "}\n",
    "\n",
    "postprocess = {\n",
    "    \"Age\": {\"type\": \"suffix\", \"rule\": \" years old\"},\n",
    "    \"Sex\": {\"type\": \"transcoding\", \"rule\": {\"male\": \"Man\", \"female\": \"Woman\"}},\n",
    "    \"Pclass\": {\"type\": \"transcoding\", \"rule\": {1: \"First\", 2: \"Second\", 3: \"Third\"}},\n",
    "}\n",
    "\n",
    "label_dict = {0: \"Deceased\", 1: \"Survived\"}\n",
    "\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7c76e52b",
   "metadata": {},
   "source": [
    "## 2. Encode and split train/test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9dce7ce0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train target distribution (initial):\n",
      "Survived\n",
      "0    0.616766\n",
      "1    0.383234\n",
      "Name: ratio, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "encoder = one_hot.OneHotEncoder(cols=[\"Sex\"])\n",
    "X_enc = encoder.fit_transform(X)\n",
    "\n",
    "X_train_full, X_test, y_train_full, y_test = train_test_split(\n",
    "    X_enc,\n",
    "    y,\n",
    "    test_size=0.25,\n",
    "    random_state=42,\n",
    "    stratify=y,\n",
    ")\n",
    "\n",
    "print(\"Train target distribution (initial):\")\n",
    "print(y_train_full[target_name].value_counts(normalize=True).rename(\"ratio\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51e2b890",
   "metadata": {},
   "source": [
    "## 3. Create an imbalanced training sample\n",
    "\n",
    "We keep all majority class examples (`Survived = 0`) and downsample the minority class (`Survived = 1`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "602e4025",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train target distribution (imbalanced):\n",
      "Survived\n",
      "0    0.871036\n",
      "1    0.128964\n",
      "Name: ratio, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "y_train_series = y_train_full[target_name]\n",
    "majority_idx = y_train_series[y_train_series == 0].index\n",
    "minority_idx = y_train_series[y_train_series == 1].index\n",
    "\n",
    "# Keep about 15% positives compared to negatives for a strong imbalance\n",
    "rng = np.random.default_rng(seed=42)\n",
    "n_minority_kept = max(1, int(0.15 * len(majority_idx)))\n",
    "n_minority_kept = min(n_minority_kept, len(minority_idx))\n",
    "minority_idx_kept = rng.choice(minority_idx.to_numpy(), size=n_minority_kept, replace=False)\n",
    "\n",
    "imbalanced_idx = np.concatenate([majority_idx.to_numpy(), minority_idx_kept])\n",
    "X_train_imb = X_train_full.loc[imbalanced_idx].sort_index()\n",
    "y_train_imb = y_train_full.loc[imbalanced_idx].sort_index()\n",
    "\n",
    "print(\"Train target distribution (imbalanced):\")\n",
    "print(y_train_imb[target_name].value_counts(normalize=True).rename(\"ratio\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "560bdb87",
   "metadata": {},
   "source": [
    "## 4. Train and evaluate baseline vs weighted model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c0224c37",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_binary_model(model, X_eval, y_eval, model_name):\n",
    "    y_true = y_eval[target_name]\n",
    "    y_pred = model.predict(X_eval)\n",
    "    return pd.Series(\n",
    "        {\n",
    "            \"model\": model_name,\n",
    "            \"accuracy\": accuracy_score(y_true, y_pred),\n",
    "            \"precision_minority\": precision_score(y_true, y_pred, pos_label=1, zero_division=0),\n",
    "            \"recall_minority\": recall_score(y_true, y_pred, pos_label=1, zero_division=0),\n",
    "            \"f1_minority\": f1_score(y_true, y_pred, pos_label=1, zero_division=0),\n",
    "        }\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7cfd1d0c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.microsoft.datawrangler.viewer.v0+json": {
       "columns": [
        {
         "name": "model",
         "rawType": "object",
         "type": "string"
        },
        {
         "name": "accuracy",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "precision_minority",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "recall_minority",
         "rawType": "float64",
         "type": "float"
        },
        {
         "name": "f1_minority",
         "rawType": "float64",
         "type": "float"
        }
       ],
       "ref": "7fcc0040-7bf8-42ba-8080-5d6ff5bead79",
       "rows": [
        [
         "baseline_imbalanced_train",
         "0.7354260089686099",
         "0.9090909090909091",
         "0.3488372093023256",
         "0.5042016806722689"
        ],
        [
         "weighted_imbalanced_train",
         "0.7399103139013453",
         "0.7058823529411765",
         "0.5581395348837209",
         "0.6233766233766234"
        ]
       ],
       "shape": {
        "columns": 4,
        "rows": 2
       }
      },
      "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>accuracy</th>\n",
       "      <th>precision_minority</th>\n",
       "      <th>recall_minority</th>\n",
       "      <th>f1_minority</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>baseline_imbalanced_train</th>\n",
       "      <td>0.735426</td>\n",
       "      <td>0.909091</td>\n",
       "      <td>0.348837</td>\n",
       "      <td>0.504202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>weighted_imbalanced_train</th>\n",
       "      <td>0.739910</td>\n",
       "      <td>0.705882</td>\n",
       "      <td>0.558140</td>\n",
       "      <td>0.623377</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           accuracy  precision_minority  recall_minority  \\\n",
       "model                                                                      \n",
       "baseline_imbalanced_train  0.735426            0.909091         0.348837   \n",
       "weighted_imbalanced_train  0.739910            0.705882         0.558140   \n",
       "\n",
       "                           f1_minority  \n",
       "model                                   \n",
       "baseline_imbalanced_train     0.504202  \n",
       "weighted_imbalanced_train     0.623377  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "baseline_model = LGBMClassifier(max_depth=3, n_estimators=300, random_state=42, verbose=-1)\n",
    "baseline_model.fit(X_train_imb, y_train_imb[target_name])\n",
    "\n",
    "neg_count = int((y_train_imb[target_name] == 0).sum())\n",
    "pos_count = int((y_train_imb[target_name] == 1).sum())\n",
    "scale_pos_weight = neg_count / pos_count\n",
    "\n",
    "weighted_model = LGBMClassifier(\n",
    "    max_depth=3,\n",
    "    n_estimators=300,\n",
    "    random_state=42,\n",
    "    scale_pos_weight=scale_pos_weight,\n",
    "    verbose=-1,\n",
    ")\n",
    "weighted_model.fit(X_train_imb, y_train_imb[target_name])\n",
    "\n",
    "metrics_baseline = evaluate_binary_model(baseline_model, X_test, y_test, \"baseline_imbalanced_train\")\n",
    "metrics_weighted = evaluate_binary_model(weighted_model, X_test, y_test, \"weighted_imbalanced_train\")\n",
    "\n",
    "comparison = pd.DataFrame([metrics_baseline, metrics_weighted]).set_index(\"model\")\n",
    "comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fe43f40a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report - weighted model\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0      0.755     0.854     0.801       137\n",
      "           1      0.706     0.558     0.623        86\n",
      "\n",
      "    accuracy                          0.740       223\n",
      "   macro avg      0.730     0.706     0.712       223\n",
      "weighted avg      0.736     0.740     0.733       223\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"Classification report - weighted model\")\n",
    "print(classification_report(y_test[target_name], weighted_model.predict(X_test), digits=3, zero_division=0))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bbfb2ec",
   "metadata": {},
   "source": [
    "## 5. Explain the weighted model with Shapash"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "45a7f47f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO: Shap explainer type - <shap.explainers._tree.TreeExplainer object at 0x11d9c05f0>\n"
     ]
    }
   ],
   "source": [
    "y_target_test = y_test.copy()\n",
    "\n",
    "additional_data = titanic_df.loc[X_test.index, [\"Name\", \"Fare\", \"Title\", \"Embarked\"]]\n",
    "additional_features_dict = {\n",
    "    \"Name\": \"Passenger Name\",\n",
    "    \"Fare\": \"Ticket Fare\",\n",
    "    \"Title\": \"Passenger Title\",\n",
    "    \"Embarked\": \"Embarkation Port\",\n",
    "}\n",
    "\n",
    "xpl = SmartExplainer(\n",
    "    model=weighted_model,\n",
    "    preprocessing=encoder,\n",
    "    postprocessing=postprocess,\n",
    "    features_dict=feature_dict,\n",
    "    label_dict=label_dict,\n",
    "    title_story=\"Titanic binary classification with imbalanced training\",\n",
    ")\n",
    "\n",
    "xpl.compile(\n",
    "    x=X_test,\n",
    "    y_target=y_target_test,\n",
    "    additional_data=additional_data,\n",
    "    additional_features_dict=additional_features_dict,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fae398b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Native Shapash confusion matrix based on y_target and y_pred passed to compile\n",
    "xpl.plot.confusion_matrix_plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "823f73bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "xpl.plot.features_importance()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "207142e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Display one local explanation\n",
    "xpl.plot.local_plot(index=X_test.index[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25b850e9",
   "metadata": {},
   "source": [
    "## 6. Launch the webapp\n",
    "\n",
    "Use the same approach as in `shapash/webapp/webapp_launch.py`.\n",
    "Run the next cell to start the local Shapash webapp."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "090f45f2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# xpl.run_app(title_story='Titanic imbalanced classification tutorial')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "survptf_312",
   "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": "3.12.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}