BSC_project.ipynb

bsc_project.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyNIWt9YZ/k7c6Y4xUNbVSYo",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/alonsosilvaallende/2a88abda7dfa28cd17b37b80cbaf95f9/bsc_project.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "lou1mDgsb4d6"
      },
      "outputs": [],
      "source": [
        "!pip install -q scikit-survival"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -q eli5"
      ],
      "metadata": {
        "id": "jjqvp6PC1PVT"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -q shap"
      ],
      "metadata": {
        "id": "FSZYrbwn6AWz"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "45lQix8lcA2y"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sksurv.datasets import load_gbsg2\n",
        "\n",
        "X, y = load_gbsg2()\n",
        "X.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "zZHRTFkdcIM4",
        "outputId": "f78f4767-6d2c-4f99-c954-93f96c069183"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    age  estrec horTh menostat  pnodes  progrec tgrade  tsize\n",
              "0  70.0    66.0    no     Post     3.0     48.0     II   21.0\n",
              "1  56.0    77.0   yes     Post     7.0     61.0     II   12.0\n",
              "2  58.0   271.0   yes     Post     9.0     52.0     II   35.0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d17a4e67-f59d-49d6-ae58-db72708f751b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>age</th>\n",
              "      <th>estrec</th>\n",
              "      <th>horTh</th>\n",
              "      <th>menostat</th>\n",
              "      <th>pnodes</th>\n",
              "      <th>progrec</th>\n",
              "      <th>tgrade</th>\n",
              "      <th>tsize</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>70.0</td>\n",
              "      <td>66.0</td>\n",
              "      <td>no</td>\n",
              "      <td>Post</td>\n",
              "      <td>3.0</td>\n",
              "      <td>48.0</td>\n",
              "      <td>II</td>\n",
              "      <td>21.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>56.0</td>\n",
              "      <td>77.0</td>\n",
              "      <td>yes</td>\n",
              "      <td>Post</td>\n",
              "      <td>7.0</td>\n",
              "      <td>61.0</td>\n",
              "      <td>II</td>\n",
              "      <td>12.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>58.0</td>\n",
              "      <td>271.0</td>\n",
              "      <td>yes</td>\n",
              "      <td>Post</td>\n",
              "      <td>9.0</td>\n",
              "      <td>52.0</td>\n",
              "      <td>II</td>\n",
              "      <td>35.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d17a4e67-f59d-49d6-ae58-db72708f751b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-d17a4e67-f59d-49d6-ae58-db72708f751b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-d17a4e67-f59d-49d6-ae58-db72708f751b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y[:3]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ep87Os_IcLI6",
        "outputId": "324b232b-be85-4c06-c885-f0e962a37dea"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([( True, 1814.), ( True, 2018.), ( True,  712.)],\n",
              "      dtype=[('cens', '?'), ('time', '<f8')])"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"% censorship: {100*(1-np.mean(y['cens'])):.2f}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IHYpMoqFcPpA",
        "outputId": "93c07c5a-ed43-4054-f864-342e9ed2b744"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "% censorship: 56.41\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_trn, X_test, y_trn, y_test = train_test_split(X, y, random_state=42)"
      ],
      "metadata": {
        "id": "c1aVxKXJccAr"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X.dtypes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "koZphqbdcl5q",
        "outputId": "83b85507-5c74-4fba-c40d-e48c4722ae9a"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "age          float64\n",
              "estrec       float64\n",
              "horTh       category\n",
              "menostat    category\n",
              "pnodes       float64\n",
              "progrec      float64\n",
              "tgrade      category\n",
              "tsize        float64\n",
              "dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def my_score(est, X_trn=X_trn, y_trn=y_trn, X_test=X_test, y_test=y_test):\n",
        "  est.fit(X_trn, y_trn)\n",
        "  ci = concordance_index_censored(y_test[\"cens\"], y_test[\"time\"], est.predict(X_test))\n",
        "  survs = est.predict_survival_function(X_test)\n",
        "  times = np.arange(365, 1826)\n",
        "  preds = np.asarray([[fn(t) for t in times] for fn in survs])\n",
        "  ibs = integrated_brier_score(y_trn, y_test, preds, times)\n",
        "  print(f'Concordance index: {ci[0]}')\n",
        "  print(f'Integrated brier score: {ibs}')"
      ],
      "metadata": {
        "id": "Ngpbzw5OdeMU"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "scaling_cols = [c for c in X.columns if X[c].dtype.kind in ['i', 'f']]\n",
        "cat_cols = [c for c in X.columns if X[c].dtype.kind not in ['i', 'f']]"
      ],
      "metadata": {
        "id": "BFZSrnvYcqGt"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.preprocessing import OrdinalEncoder\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "preprocessor = ColumnTransformer(\n",
        "    [('cat-preprocessor', OrdinalEncoder(), cat_cols),\n",
        "    ('standard-scaler', StandardScaler(), scaling_cols)],\n",
        "    remainder='passthrough', sparse_threshold=0)"
      ],
      "metadata": {
        "id": "FwVbgZavcunZ"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.pipeline import make_pipeline\n",
        "from sksurv.linear_model import CoxPHSurvivalAnalysis\n",
        "from sksurv.metrics import concordance_index_censored\n",
        "from sksurv.metrics import integrated_brier_score\n",
        "\n",
        "cph = make_pipeline(preprocessor, CoxPHSurvivalAnalysis())\n",
        "my_score(cph)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i0mE9J_XcyAl",
        "outputId": "02e04bc4-d9c9-4ee6-c93e-92c54c5f7520"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.634554233894712\n",
            "Integrated brier score: 0.20009274951127964\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "\n",
        "param_grid = {\n",
        "    'coxphsurvivalanalysis__alpha': 10**np.linspace(-1,2,5),\n",
        "}\n",
        "cph_gs = GridSearchCV(\n",
        "    cph, param_grid=param_grid, n_jobs=-1, cv=3)\n",
        "cph_gs.fit(X_trn, y_trn)\n",
        "cph_best = make_pipeline(preprocessor, CoxPHSurvivalAnalysis(alpha=cph_gs.best_params_['coxphsurvivalanalysis__alpha']))\n",
        "cph_best.fit(X_trn, y_trn)\n",
        "my_score(cph_best)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l3hsTAtne71k",
        "outputId": "ecd3a9dc-0600-4f08-b04c-1f6368e729df"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.6430338004946414\n",
            "Integrated brier score: 0.19743464033866906\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sksurv.ensemble import RandomSurvivalForest\n",
        "\n",
        "rsf = make_pipeline(preprocessor, RandomSurvivalForest(random_state=42))\n",
        "rsf.fit(X_trn, y_trn)\n",
        "my_score(rsf)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hlj--y94u-as",
        "outputId": "51da1c55-127b-4869-e42b-256699bad9ce"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.6579908138028501\n",
            "Integrated brier score: 0.1984125909087399\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rsf.get_params()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "B_dFO8mYvOOm",
        "outputId": "ff6f4935-cd86-44cf-917a-5e2e0e240f6b"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'memory': None,\n",
              " 'steps': [('columntransformer',\n",
              "   ColumnTransformer(remainder='passthrough', sparse_threshold=0,\n",
              "                     transformers=[('cat-preprocessor', OrdinalEncoder(),\n",
              "                                    ['horTh', 'menostat', 'tgrade']),\n",
              "                                   ('standard-scaler', StandardScaler(),\n",
              "                                    ['age', 'estrec', 'pnodes', 'progrec',\n",
              "                                     'tsize'])])),\n",
              "  ('randomsurvivalforest', RandomSurvivalForest(random_state=42))],\n",
              " 'verbose': False,\n",
              " 'columntransformer': ColumnTransformer(remainder='passthrough', sparse_threshold=0,\n",
              "                   transformers=[('cat-preprocessor', OrdinalEncoder(),\n",
              "                                  ['horTh', 'menostat', 'tgrade']),\n",
              "                                 ('standard-scaler', StandardScaler(),\n",
              "                                  ['age', 'estrec', 'pnodes', 'progrec',\n",
              "                                   'tsize'])]),\n",
              " 'randomsurvivalforest': RandomSurvivalForest(random_state=42),\n",
              " 'columntransformer__n_jobs': None,\n",
              " 'columntransformer__remainder': 'passthrough',\n",
              " 'columntransformer__sparse_threshold': 0,\n",
              " 'columntransformer__transformer_weights': None,\n",
              " 'columntransformer__transformers': [('cat-preprocessor',\n",
              "   OrdinalEncoder(),\n",
              "   ['horTh', 'menostat', 'tgrade']),\n",
              "  ('standard-scaler',\n",
              "   StandardScaler(),\n",
              "   ['age', 'estrec', 'pnodes', 'progrec', 'tsize'])],\n",
              " 'columntransformer__verbose': False,\n",
              " 'columntransformer__verbose_feature_names_out': True,\n",
              " 'columntransformer__cat-preprocessor': OrdinalEncoder(),\n",
              " 'columntransformer__standard-scaler': StandardScaler(),\n",
              " 'columntransformer__cat-preprocessor__categories': 'auto',\n",
              " 'columntransformer__cat-preprocessor__dtype': numpy.float64,\n",
              " 'columntransformer__cat-preprocessor__handle_unknown': 'error',\n",
              " 'columntransformer__cat-preprocessor__unknown_value': None,\n",
              " 'columntransformer__standard-scaler__copy': True,\n",
              " 'columntransformer__standard-scaler__with_mean': True,\n",
              " 'columntransformer__standard-scaler__with_std': True,\n",
              " 'randomsurvivalforest__bootstrap': True,\n",
              " 'randomsurvivalforest__max_depth': None,\n",
              " 'randomsurvivalforest__max_features': 'auto',\n",
              " 'randomsurvivalforest__max_leaf_nodes': None,\n",
              " 'randomsurvivalforest__max_samples': None,\n",
              " 'randomsurvivalforest__min_samples_leaf': 3,\n",
              " 'randomsurvivalforest__min_samples_split': 6,\n",
              " 'randomsurvivalforest__min_weight_fraction_leaf': 0.0,\n",
              " 'randomsurvivalforest__n_estimators': 100,\n",
              " 'randomsurvivalforest__n_jobs': None,\n",
              " 'randomsurvivalforest__oob_score': False,\n",
              " 'randomsurvivalforest__random_state': 42,\n",
              " 'randomsurvivalforest__verbose': 0,\n",
              " 'randomsurvivalforest__warm_start': False}"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "param_grid_rsf = {\n",
        "    'randomsurvivalforest__max_features': np.arange(3, 8),\n",
        "    'randomsurvivalforest__max_depth': [5,10,None],\n",
        "    'randomsurvivalforest__min_samples_leaf': [1,3,5],\n",
        "}\n",
        "\n",
        "rsf_gs = GridSearchCV(\n",
        "    rsf, param_grid=param_grid_rsf, cv=3)\n",
        "rsf_gs.fit(X_trn, y_trn)\n",
        "rsf_best = make_pipeline(preprocessor, RandomSurvivalForest(max_depth= rsf_gs.best_params_['randomsurvivalforest__max_depth'],\n",
        "                     max_features= rsf_gs.best_params_['randomsurvivalforest__max_features'],\n",
        "                     min_samples_leaf= rsf_gs.best_params_['randomsurvivalforest__min_samples_leaf']))\n",
        "rsf_best.fit(X_trn, y_trn)\n",
        "my_score(rsf_best)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hjy0rYCyvchH",
        "outputId": "5af669d2-eaa9-4181-8ed2-51c59f54c869"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.6247791779531269\n",
            "Integrated brier score: 0.20951841008965424\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sksurv.ensemble import GradientBoostingSurvivalAnalysis\n",
        "\n",
        "gbc = make_pipeline(preprocessor, GradientBoostingSurvivalAnalysis(random_state=42))\n",
        "gbc.fit(X_trn, y_trn)\n",
        "my_score(gbc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QBNY5WXjxHMb",
        "outputId": "6d7d8ab0-32dc-488a-e187-f674f6aa5824"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.6410905664821576\n",
            "Integrated brier score: 0.19772913228546876\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "param_grid = {\n",
        "    'gradientboostingsurvivalanalysis__learning_rate': 10**np.linspace(-1,0,3),\n",
        "    'gradientboostingsurvivalanalysis__max_depth': np.arange(3, 10, 2),\n",
        "    'gradientboostingsurvivalanalysis__min_samples_leaf': np.arange(2,6,2),\n",
        "}\n",
        "\n",
        "gbc_gs = GridSearchCV(gbc, param_grid=param_grid, cv=3, n_jobs=-1)\n",
        "gbc_gs.fit(X_trn, y_trn)\n",
        "gbc_best = make_pipeline(preprocessor, GradientBoostingSurvivalAnalysis(learning_rate= gbc_gs.best_params_['gradientboostingsurvivalanalysis__learning_rate'],\n",
        "                     max_depth= gbc_gs.best_params_['gradientboostingsurvivalanalysis__max_depth'],\n",
        "                     min_samples_leaf= gbc_gs.best_params_['gradientboostingsurvivalanalysis__min_samples_leaf']))\n",
        "gbc_best.fit(X_trn, y_trn)\n",
        "my_score(gbc_best)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AuKdh16qxvqo",
        "outputId": "7488dccc-8bf0-4de2-aada-875d130be4cf"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/joblib/externals/loky/process_executor.py:703: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n",
            "  \"timeout or by a memory leak.\", UserWarning\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.6283123307030974\n",
            "Integrated brier score: 0.20318280691517193\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from eli5.sklearn import PermutationImportance"
      ],
      "metadata": {
        "id": "EcS_gT830kw5"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "perm = PermutationImportance(\n",
        "    rsf.steps[-1][1], n_iter=100, random_state=42).fit(preprocessor.fit_transform(X_test),y_test)"
      ],
      "metadata": {
        "id": "NYZzSt_V0lN7"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import eli5\n",
        "eli5.show_weights(perm, feature_names = X.columns.tolist())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        },
        "id": "F9p0vI4t0nvn",
        "outputId": "0387a255-9f51-48cb-fab3-98c95a72acb1"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <style>\n",
              "    table.eli5-weights tr:hover {\n",
              "        filter: brightness(85%);\n",
              "    }\n",
              "</style>\n",
              "\n",
              "\n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "        <table class=\"eli5-weights eli5-feature-importances\" style=\"border-collapse: collapse; border: none; margin-top: 0em; table-layout: auto;\">\n",
              "    <thead>\n",
              "    <tr style=\"border: none;\">\n",
              "        <th style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">Weight</th>\n",
              "        <th style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">Feature</th>\n",
              "    </tr>\n",
              "    </thead>\n",
              "    <tbody>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(120, 100.00%, 80.00%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                0.0886\n",
              "                \n",
              "                    &plusmn; 0.0426\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                progrec\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(120, 100.00%, 92.58%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                0.0215\n",
              "                \n",
              "                    &plusmn; 0.0417\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                tgrade\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(120, 100.00%, 95.39%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                0.0109\n",
              "                \n",
              "                    &plusmn; 0.0197\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                menostat\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(120, 100.00%, 95.69%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                0.0099\n",
              "                \n",
              "                    &plusmn; 0.0111\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                age\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(120, 100.00%, 96.31%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                0.0079\n",
              "                \n",
              "                    &plusmn; 0.0112\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                horTh\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(120, 100.00%, 96.36%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                0.0078\n",
              "                \n",
              "                    &plusmn; 0.0197\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                pnodes\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(0, 100.00%, 97.96%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                -0.0034\n",
              "                \n",
              "                    &plusmn; 0.0065\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                estrec\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "        <tr style=\"background-color: hsl(0, 100.00%, 95.52%); border: none;\">\n",
              "            <td style=\"padding: 0 1em 0 0.5em; text-align: right; border: none;\">\n",
              "                -0.0105\n",
              "                \n",
              "                    &plusmn; 0.0186\n",
              "                \n",
              "            </td>\n",
              "            <td style=\"padding: 0 0.5em 0 0.5em; text-align: left; border: none;\">\n",
              "                tsize\n",
              "            </td>\n",
              "        </tr>\n",
              "    \n",
              "    \n",
              "    </tbody>\n",
              "</table>\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "    \n",
              "\n",
              "\n",
              "\n"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data = perm.results_\n",
        "data = pd.DataFrame(data, columns=X_trn.columns)\n",
        "meds = data.median()\n",
        "meds = meds.sort_values(ascending=False)\n",
        "data = data[meds.index]\n",
        "fig, ax = plt.subplots()\n",
        "data.boxplot(ax=ax)\n",
        "ax.set_title('Permutation Importances')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "dYKh6YRq1afe",
        "outputId": "f3d77340-e10c-4045-febc-c600b617fd7c"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.feature_selection import SelectFromModel\n",
        "\n",
        "sel = SelectFromModel(perm, max_features=6, prefit=True)\n",
        "X_trn_trans = sel.transform(X_trn)\n",
        "X_test_trans = sel.transform(X_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-nyBcGgW1d7I",
        "outputId": "38fee0f6-3142-4603-f9eb-ca61b85517d1"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/base.py:444: UserWarning: X has feature names, but SelectFromModel was fitted without feature names\n",
            "  f\"X has feature names, but {self.__class__.__name__} was fitted without\"\n",
            "/usr/local/lib/python3.7/dist-packages/sklearn/base.py:444: UserWarning: X has feature names, but SelectFromModel was fitted without feature names\n",
            "  f\"X has feature names, but {self.__class__.__name__} was fitted without\"\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X.columns[sel.get_support()]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0zejgr1b2uOo",
        "outputId": "b148c566-87ca-424b-8895-56eae2cb7e22"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['progrec', 'tgrade'], dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "scaling_cols = [c for c in X.columns[sel.get_support()] if X[c].dtype.kind in ['i', 'f']]\n",
        "cat_cols = [c for c in X.columns[sel.get_support()] if X[c].dtype.kind not in ['i', 'f']]"
      ],
      "metadata": {
        "id": "v0X44JiO35mj"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "preprocessor = ColumnTransformer(\n",
        "    [('cat-preprocessor', OrdinalEncoder(), cat_cols),\n",
        "    ('standard-scaler', StandardScaler(), scaling_cols)],\n",
        "    remainder='passthrough', sparse_threshold=0)"
      ],
      "metadata": {
        "id": "USgbwaUT4eRj"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rsf = make_pipeline(preprocessor, RandomSurvivalForest(random_state=42))\n",
        "rsf.fit(pd.DataFrame(X_trn_trans, columns=X.columns[sel.get_support()]), y_trn)\n",
        "my_score(rsf, pd.DataFrame(X_trn_trans, columns=X.columns[sel.get_support()]), y_trn, pd.DataFrame(X_test_trans, columns=X.columns[sel.get_support()]), y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5iF9aF6J1lYD",
        "outputId": "2da908fa-b772-486e-ce68-6c8b764338ef"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Concordance index: 0.557237074549523\n",
            "Integrated brier score: 0.2407429762535294\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import shap"
      ],
      "metadata": {
        "id": "kzfBU7ei6abi"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "cat_cols = [c for c in X.columns if X[c].dtype.kind not in ['i', 'f'] and c != \"tgrade\"]"
      ],
      "metadata": {
        "id": "TnwCnwm1XBsR"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import OneHotEncoder\n",
        "\n",
        "preprocessor = ColumnTransformer(\n",
        "    [('cat-preprocessor', OneHotEncoder(drop='first'), cat_cols)],\n",
        "    remainder='passthrough', sparse_threshold=0)"
      ],
      "metadata": {
        "id": "AZF1KRX8XKEf"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "grade_str = X[\"tgrade\"].astype(object).values[:, None]\n",
        "grade_num = OrdinalEncoder(categories=[[\"I\", \"II\", \"III\"]]).fit_transform(grade_str)\n",
        "X_no_grade = X.drop(columns=\"tgrade\")\n",
        "Xt = pd.DataFrame(preprocessor.fit_transform(X_no_grade), columns=preprocessor.get_feature_names_out())\n",
        "Xt[\"tgrade\"] = grade_num"
      ],
      "metadata": {
        "id": "yfgfqi5CUca_"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_trn1, X_test1, y_trn1, y_test1 = train_test_split(Xt, y, random_state=42)"
      ],
      "metadata": {
        "id": "SOGfyRo9PzbU"
      },
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X100 = shap.utils.sample(X_trn1, 100)"
      ],
      "metadata": {
        "id": "BwuErm8yQEhC"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rsf = RandomSurvivalForest(random_state=42)\n",
        "rsf.fit(X_trn1, y_trn1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C2OumfIFQMYU",
        "outputId": "14b94a03-3b66-4ea2-9dd8-c0c289a14e98"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomSurvivalForest(random_state=42)"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# compute the SHAP values\n",
        "explainer = shap.Explainer(rsf.predict, X100)"
      ],
      "metadata": {
        "id": "rtZhm9uEQRyV"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "shap_values = explainer(X_test1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "blTnZQxZNm1-",
        "outputId": "43afc93f-eecb-49f9-8f0f-2c73f06ed3bd"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Exact explainer: 173it [05:47,  2.06s/it]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "shap.plots.waterfall(shap_values[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 389
        },
        "id": "IPiTWws_Li-z",
        "outputId": "d3ebe870-58a7-4b6d-d6e7-877e0da29293"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x396 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "shap.plots.beeswarm(shap_values)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 324
        },
        "id": "NhlkwF3yLxMW",
        "outputId": "ba109969-ed30-4d38-b0f9-d3ac10415ea3"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x338.4 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}