examples/Pycox_Example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# Evaluation for Pycox model\n",
    "\n",
    "In this notebook we introduce the use of `PycoxEvaluator` in the package through an example dataset and a model.\n",
    "The first half of this notebook follows the procedure described in the Pycox [example](https://github.com/havakv/pycox/blob/master/examples/01_introduction.ipynb).\n",
    "\n",
    "In the following we will:\n",
    "\n",
    "1. Load the METABRIC survival dataset and create a `LogisticHazard` model;\n",
    "2. Train the model using training set;\n",
    "3. Evaluate the performance using the concordance index, Brier score, Integrated Brier score, L1 loss, 1-calibration, and D-calibration."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": [
    "!pip install pycox\n",
    "!pip install scikit_pandas\n",
    "!pip install torchtuples"
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Import Dependencies\n",
    "Note: You need `sklearn-pandas` for normalizing the data."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:45.189338Z",
     "start_time": "2024-04-28T21:03:43.908163Z"
    }
   },
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# For preprocessing\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn_pandas import DataFrameMapper\n",
    "\n",
    "import torch # For building the networks\n",
    "import torchtuples as tt # Some useful functions\n",
    "\n",
    "from pycox.datasets import metabric\n",
    "from pycox.models import LogisticHazard\n",
    "\n",
    "from Evaluator import PycoxEvaluator"
   ],
   "outputs": [],
   "execution_count": 1
  },
  {
   "cell_type": "code",
   "source": [
    "%matplotlib inline"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:46.454965Z",
     "start_time": "2024-04-28T21:03:46.449673Z"
    }
   },
   "outputs": [],
   "execution_count": 2
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Load Dataset\n",
    "Set seed for torch and numpy for reproducibility"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "np.random.seed(1234)\n",
    "_ = torch.manual_seed(123)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:48.154882Z",
     "start_time": "2024-04-28T21:03:48.152496Z"
    }
   },
   "outputs": [],
   "execution_count": 3
  },
  {
   "cell_type": "markdown",
   "source": [
    "Load the dataset, split it into training (64%), validation (16%) and testing set (20%).\n",
    "Then standardize the covariates."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "df_train = metabric.read_df()\n",
    "df_test = df_train.sample(frac=0.2)\n",
    "df_train = df_train.drop(df_test.index)\n",
    "df_val = df_train.sample(frac=0.2)\n",
    "df_train = df_train.drop(df_val.index)\n",
    "\n",
    "cols_standardize = ['x0', 'x1', 'x2', 'x3', 'x8']\n",
    "cols_leave = ['x4', 'x5', 'x6', 'x7']\n",
    "\n",
    "standardize = [([col], StandardScaler()) for col in cols_standardize]\n",
    "leave = [(col, None) for col in cols_leave]\n",
    "\n",
    "x_mapper = DataFrameMapper(standardize + leave)\n",
    "\n",
    "x_train = x_mapper.fit_transform(df_train).astype('float32')\n",
    "x_val = x_mapper.transform(df_val).astype('float32')\n",
    "x_test = x_mapper.transform(df_test).astype('float32')"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:49.828426Z",
     "start_time": "2024-04-28T21:03:49.810365Z"
    }
   },
   "outputs": [],
   "execution_count": 4
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Build the model and train it."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "num_durations = 10\n",
    "\n",
    "labtrans = LogisticHazard.label_transform(num_durations)\n",
    "\n",
    "get_target = lambda df: (df['duration'].values, df['event'].values)\n",
    "y_train = labtrans.fit_transform(*get_target(df_train))\n",
    "y_val = labtrans.transform(*get_target(df_val))\n",
    "\n",
    "train = (x_train, y_train)\n",
    "val = (x_val, y_val)\n",
    "\n",
    "# We don't need to transform the test labels\n",
    "durations_test, events_test = get_target(df_test)\n",
    "durations_train, events_train = get_target(df_train)\n",
    "\n",
    "in_features = x_train.shape[1]\n",
    "num_nodes = [32, 32]\n",
    "out_features = labtrans.out_features\n",
    "batch_norm = True\n",
    "dropout = 0.1\n",
    "\n",
    "net = tt.practical.MLPVanilla(in_features, num_nodes, out_features, batch_norm, dropout)\n",
    "\n",
    "model = LogisticHazard(net, tt.optim.Adam(0.01), duration_index=labtrans.cuts)\n",
    "\n",
    "batch_size = 256\n",
    "epochs = 100\n",
    "callbacks = [tt.cb.EarlyStopping()]\n",
    "\n",
    "log = model.fit(x_train, y_train, batch_size, epochs, callbacks, val_data=val)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:53.042455Z",
     "start_time": "2024-04-28T21:03:50.969598Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\t[0s / 0s],\t\ttrain_loss: 2.9332,\tval_loss: 2.7593\n",
      "1:\t[0s / 0s],\t\ttrain_loss: 2.6493,\tval_loss: 2.5245\n",
      "2:\t[0s / 0s],\t\ttrain_loss: 2.3807,\tval_loss: 2.2181\n",
      "3:\t[0s / 0s],\t\ttrain_loss: 2.0595,\tval_loss: 1.8908\n",
      "4:\t[0s / 0s],\t\ttrain_loss: 1.7514,\tval_loss: 1.6019\n",
      "5:\t[0s / 0s],\t\ttrain_loss: 1.5335,\tval_loss: 1.4371\n",
      "6:\t[0s / 0s],\t\ttrain_loss: 1.4379,\tval_loss: 1.3759\n",
      "7:\t[0s / 0s],\t\ttrain_loss: 1.3954,\tval_loss: 1.3658\n",
      "8:\t[0s / 0s],\t\ttrain_loss: 1.3906,\tval_loss: 1.3619\n",
      "9:\t[0s / 0s],\t\ttrain_loss: 1.3730,\tval_loss: 1.3531\n",
      "10:\t[0s / 0s],\t\ttrain_loss: 1.3605,\tval_loss: 1.3486\n",
      "11:\t[0s / 0s],\t\ttrain_loss: 1.3349,\tval_loss: 1.3513\n",
      "12:\t[0s / 0s],\t\ttrain_loss: 1.3153,\tval_loss: 1.3488\n",
      "13:\t[0s / 0s],\t\ttrain_loss: 1.3133,\tval_loss: 1.3418\n",
      "14:\t[0s / 0s],\t\ttrain_loss: 1.2988,\tval_loss: 1.3432\n",
      "15:\t[0s / 0s],\t\ttrain_loss: 1.3079,\tval_loss: 1.3525\n",
      "16:\t[0s / 0s],\t\ttrain_loss: 1.2976,\tval_loss: 1.3565\n",
      "17:\t[0s / 0s],\t\ttrain_loss: 1.2918,\tval_loss: 1.3577\n",
      "18:\t[0s / 0s],\t\ttrain_loss: 1.2900,\tval_loss: 1.3594\n",
      "19:\t[0s / 0s],\t\ttrain_loss: 1.2888,\tval_loss: 1.3655\n",
      "20:\t[0s / 0s],\t\ttrain_loss: 1.2871,\tval_loss: 1.3675\n",
      "21:\t[0s / 0s],\t\ttrain_loss: 1.2807,\tval_loss: 1.3648\n",
      "22:\t[0s / 0s],\t\ttrain_loss: 1.2920,\tval_loss: 1.3633\n",
      "23:\t[0s / 0s],\t\ttrain_loss: 1.2735,\tval_loss: 1.3681\n"
     ]
    }
   ],
   "execution_count": 5
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Make the prediction.\n",
    "`surv` is a Dataframe object. Row index represents the time coordinates, and each column represents a survival curve of one patient."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "surv = model.interpolate(10).predict_surv_df(x_test)\n",
    "surv.head(20)"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:53.867462Z",
     "start_time": "2024-04-28T21:03:53.842389Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "                0         1         2         3         4         5    \\\n",
       "0.000000   0.997888  0.998990  0.977142  0.996769  0.996701  0.997470   \n",
       "3.946667   0.990678  0.977079  0.967224  0.988112  0.975188  0.992005   \n",
       "7.893333   0.983467  0.955168  0.957305  0.979455  0.953674  0.986539   \n",
       "11.840000  0.976257  0.933257  0.947387  0.970799  0.932161  0.981074   \n",
       "15.786667  0.969046  0.911346  0.937468  0.962142  0.910647  0.975609   \n",
       "19.733334  0.961836  0.889435  0.927550  0.953485  0.889133  0.970144   \n",
       "23.680000  0.954625  0.867524  0.917631  0.944828  0.867620  0.964679   \n",
       "27.626667  0.947415  0.845613  0.907713  0.936171  0.846106  0.959213   \n",
       "31.573334  0.940204  0.823702  0.897794  0.927515  0.824592  0.953748   \n",
       "35.520000  0.932994  0.801792  0.887876  0.918858  0.803079  0.948283   \n",
       "39.466667  0.925783  0.779881  0.877957  0.910201  0.781565  0.942818   \n",
       "43.413334  0.902162  0.762350  0.862645  0.897250  0.754873  0.931251   \n",
       "47.360001  0.878541  0.744820  0.847333  0.884298  0.728181  0.919684   \n",
       "51.306667  0.854920  0.727290  0.832021  0.871347  0.701489  0.908117   \n",
       "55.253334  0.831299  0.709760  0.816709  0.858395  0.674797  0.896550   \n",
       "59.200001  0.807678  0.692230  0.801397  0.845444  0.648105  0.884983   \n",
       "63.146667  0.784057  0.674700  0.786085  0.832492  0.621412  0.873416   \n",
       "67.093334  0.760436  0.657170  0.770774  0.819541  0.594720  0.861849   \n",
       "71.040001  0.736815  0.639639  0.755462  0.806590  0.568028  0.850281   \n",
       "74.986668  0.713194  0.622109  0.740150  0.793638  0.541336  0.838714   \n",
       "\n",
       "                6         7         8         9    ...       371       372  \\\n",
       "0.000000   0.998884  0.999274  0.999061  0.996440  ...  0.994224  0.994715   \n",
       "3.946667   0.997279  0.998690  0.972076  0.938016  ...  0.982216  0.967455   \n",
       "7.893333   0.995674  0.998106  0.945092  0.879592  ...  0.970207  0.940196   \n",
       "11.840000  0.994069  0.997523  0.918107  0.821168  ...  0.958198  0.912936   \n",
       "15.786667  0.992463  0.996939  0.891123  0.762744  ...  0.946189  0.885676   \n",
       "19.733334  0.990858  0.996355  0.864138  0.704320  ...  0.934180  0.858417   \n",
       "23.680000  0.989253  0.995771  0.837153  0.645896  ...  0.922171  0.831157   \n",
       "27.626667  0.987647  0.995187  0.810169  0.587472  ...  0.910162  0.803897   \n",
       "31.573334  0.986042  0.994603  0.783184  0.529048  ...  0.898153  0.776638   \n",
       "35.520000  0.984437  0.994019  0.756200  0.470624  ...  0.886145  0.749378   \n",
       "39.466667  0.982831  0.993435  0.729215  0.412200  ...  0.874136  0.722118   \n",
       "43.413334  0.978628  0.991024  0.685955  0.395379  ...  0.863165  0.693594   \n",
       "47.360001  0.974425  0.988613  0.642695  0.378557  ...  0.852194  0.665070   \n",
       "51.306667  0.970222  0.986202  0.599434  0.361736  ...  0.841223  0.636545   \n",
       "55.253334  0.966018  0.983791  0.556174  0.344915  ...  0.830252  0.608021   \n",
       "59.200001  0.961815  0.981379  0.512914  0.328093  ...  0.819281  0.579497   \n",
       "63.146667  0.957612  0.978968  0.469653  0.311272  ...  0.808310  0.550972   \n",
       "67.093334  0.953409  0.976557  0.426393  0.294451  ...  0.797339  0.522448   \n",
       "71.040001  0.949205  0.974146  0.383133  0.277629  ...  0.786368  0.493924   \n",
       "74.986668  0.945002  0.971735  0.339872  0.260808  ...  0.775397  0.465400   \n",
       "\n",
       "                373       374       375       376       377       378  \\\n",
       "0.000000   0.996657  0.997932  0.996396  0.995586  0.997912  0.999046   \n",
       "3.946667   0.982801  0.995717  0.985402  0.985974  0.988225  0.964453   \n",
       "7.893333   0.968944  0.993502  0.974408  0.976363  0.978538  0.929859   \n",
       "11.840000  0.955088  0.991287  0.963414  0.966751  0.968852  0.895265   \n",
       "15.786667  0.941231  0.989071  0.952420  0.957139  0.959165  0.860672   \n",
       "19.733334  0.927375  0.986856  0.941425  0.947527  0.949479  0.826078   \n",
       "23.680000  0.913518  0.984641  0.930431  0.937916  0.939792  0.791484   \n",
       "27.626667  0.899662  0.982426  0.919437  0.928304  0.930105  0.756891   \n",
       "31.573334  0.885805  0.980211  0.908443  0.918692  0.920419  0.722297   \n",
       "35.520000  0.871949  0.977995  0.897449  0.909080  0.910732  0.687704   \n",
       "39.466667  0.858092  0.975780  0.886455  0.899469  0.901046  0.653110   \n",
       "43.413334  0.843046  0.970013  0.866816  0.889184  0.879391  0.608127   \n",
       "47.360001  0.828001  0.964245  0.847177  0.878898  0.857737  0.563145   \n",
       "51.306667  0.812955  0.958478  0.827538  0.868613  0.836082  0.518162   \n",
       "55.253334  0.797909  0.952711  0.807900  0.858328  0.814428  0.473180   \n",
       "59.200001  0.782863  0.946943  0.788261  0.848043  0.792773  0.428197   \n",
       "63.146667  0.767817  0.941176  0.768622  0.837758  0.771119  0.383215   \n",
       "67.093334  0.752772  0.935408  0.748983  0.827473  0.749465  0.338232   \n",
       "71.040001  0.737726  0.929641  0.729344  0.817188  0.727810  0.293250   \n",
       "74.986668  0.722680  0.923874  0.709706  0.806903  0.706156  0.248267   \n",
       "\n",
       "                379       380  \n",
       "0.000000   0.998036  0.987776  \n",
       "3.946667   0.974502  0.965598  \n",
       "7.893333   0.950968  0.943420  \n",
       "11.840000  0.927434  0.921242  \n",
       "15.786667  0.903900  0.899064  \n",
       "19.733334  0.880365  0.876886  \n",
       "23.680000  0.856831  0.854708  \n",
       "27.626667  0.833297  0.832530  \n",
       "31.573334  0.809763  0.810352  \n",
       "35.520000  0.786229  0.788174  \n",
       "39.466667  0.762695  0.765996  \n",
       "43.413334  0.742316  0.744343  \n",
       "47.360001  0.721937  0.722690  \n",
       "51.306667  0.701558  0.701038  \n",
       "55.253334  0.681179  0.679385  \n",
       "59.200001  0.660800  0.657733  \n",
       "63.146667  0.640421  0.636080  \n",
       "67.093334  0.620041  0.614428  \n",
       "71.040001  0.599662  0.592775  \n",
       "74.986668  0.579283  0.571123  \n",
       "\n",
       "[20 rows x 381 columns]"
      ],
      "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>371</th>\n",
       "      <th>372</th>\n",
       "      <th>373</th>\n",
       "      <th>374</th>\n",
       "      <th>375</th>\n",
       "      <th>376</th>\n",
       "      <th>377</th>\n",
       "      <th>378</th>\n",
       "      <th>379</th>\n",
       "      <th>380</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.000000</th>\n",
       "      <td>0.997888</td>\n",
       "      <td>0.998990</td>\n",
       "      <td>0.977142</td>\n",
       "      <td>0.996769</td>\n",
       "      <td>0.996701</td>\n",
       "      <td>0.997470</td>\n",
       "      <td>0.998884</td>\n",
       "      <td>0.999274</td>\n",
       "      <td>0.999061</td>\n",
       "      <td>0.996440</td>\n",
       "      <td>...</td>\n",
       "      <td>0.994224</td>\n",
       "      <td>0.994715</td>\n",
       "      <td>0.996657</td>\n",
       "      <td>0.997932</td>\n",
       "      <td>0.996396</td>\n",
       "      <td>0.995586</td>\n",
       "      <td>0.997912</td>\n",
       "      <td>0.999046</td>\n",
       "      <td>0.998036</td>\n",
       "      <td>0.987776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.946667</th>\n",
       "      <td>0.990678</td>\n",
       "      <td>0.977079</td>\n",
       "      <td>0.967224</td>\n",
       "      <td>0.988112</td>\n",
       "      <td>0.975188</td>\n",
       "      <td>0.992005</td>\n",
       "      <td>0.997279</td>\n",
       "      <td>0.998690</td>\n",
       "      <td>0.972076</td>\n",
       "      <td>0.938016</td>\n",
       "      <td>...</td>\n",
       "      <td>0.982216</td>\n",
       "      <td>0.967455</td>\n",
       "      <td>0.982801</td>\n",
       "      <td>0.995717</td>\n",
       "      <td>0.985402</td>\n",
       "      <td>0.985974</td>\n",
       "      <td>0.988225</td>\n",
       "      <td>0.964453</td>\n",
       "      <td>0.974502</td>\n",
       "      <td>0.965598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.893333</th>\n",
       "      <td>0.983467</td>\n",
       "      <td>0.955168</td>\n",
       "      <td>0.957305</td>\n",
       "      <td>0.979455</td>\n",
       "      <td>0.953674</td>\n",
       "      <td>0.986539</td>\n",
       "      <td>0.995674</td>\n",
       "      <td>0.998106</td>\n",
       "      <td>0.945092</td>\n",
       "      <td>0.879592</td>\n",
       "      <td>...</td>\n",
       "      <td>0.970207</td>\n",
       "      <td>0.940196</td>\n",
       "      <td>0.968944</td>\n",
       "      <td>0.993502</td>\n",
       "      <td>0.974408</td>\n",
       "      <td>0.976363</td>\n",
       "      <td>0.978538</td>\n",
       "      <td>0.929859</td>\n",
       "      <td>0.950968</td>\n",
       "      <td>0.943420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11.840000</th>\n",
       "      <td>0.976257</td>\n",
       "      <td>0.933257</td>\n",
       "      <td>0.947387</td>\n",
       "      <td>0.970799</td>\n",
       "      <td>0.932161</td>\n",
       "      <td>0.981074</td>\n",
       "      <td>0.994069</td>\n",
       "      <td>0.997523</td>\n",
       "      <td>0.918107</td>\n",
       "      <td>0.821168</td>\n",
       "      <td>...</td>\n",
       "      <td>0.958198</td>\n",
       "      <td>0.912936</td>\n",
       "      <td>0.955088</td>\n",
       "      <td>0.991287</td>\n",
       "      <td>0.963414</td>\n",
       "      <td>0.966751</td>\n",
       "      <td>0.968852</td>\n",
       "      <td>0.895265</td>\n",
       "      <td>0.927434</td>\n",
       "      <td>0.921242</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15.786667</th>\n",
       "      <td>0.969046</td>\n",
       "      <td>0.911346</td>\n",
       "      <td>0.937468</td>\n",
       "      <td>0.962142</td>\n",
       "      <td>0.910647</td>\n",
       "      <td>0.975609</td>\n",
       "      <td>0.992463</td>\n",
       "      <td>0.996939</td>\n",
       "      <td>0.891123</td>\n",
       "      <td>0.762744</td>\n",
       "      <td>...</td>\n",
       "      <td>0.946189</td>\n",
       "      <td>0.885676</td>\n",
       "      <td>0.941231</td>\n",
       "      <td>0.989071</td>\n",
       "      <td>0.952420</td>\n",
       "      <td>0.957139</td>\n",
       "      <td>0.959165</td>\n",
       "      <td>0.860672</td>\n",
       "      <td>0.903900</td>\n",
       "      <td>0.899064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19.733334</th>\n",
       "      <td>0.961836</td>\n",
       "      <td>0.889435</td>\n",
       "      <td>0.927550</td>\n",
       "      <td>0.953485</td>\n",
       "      <td>0.889133</td>\n",
       "      <td>0.970144</td>\n",
       "      <td>0.990858</td>\n",
       "      <td>0.996355</td>\n",
       "      <td>0.864138</td>\n",
       "      <td>0.704320</td>\n",
       "      <td>...</td>\n",
       "      <td>0.934180</td>\n",
       "      <td>0.858417</td>\n",
       "      <td>0.927375</td>\n",
       "      <td>0.986856</td>\n",
       "      <td>0.941425</td>\n",
       "      <td>0.947527</td>\n",
       "      <td>0.949479</td>\n",
       "      <td>0.826078</td>\n",
       "      <td>0.880365</td>\n",
       "      <td>0.876886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23.680000</th>\n",
       "      <td>0.954625</td>\n",
       "      <td>0.867524</td>\n",
       "      <td>0.917631</td>\n",
       "      <td>0.944828</td>\n",
       "      <td>0.867620</td>\n",
       "      <td>0.964679</td>\n",
       "      <td>0.989253</td>\n",
       "      <td>0.995771</td>\n",
       "      <td>0.837153</td>\n",
       "      <td>0.645896</td>\n",
       "      <td>...</td>\n",
       "      <td>0.922171</td>\n",
       "      <td>0.831157</td>\n",
       "      <td>0.913518</td>\n",
       "      <td>0.984641</td>\n",
       "      <td>0.930431</td>\n",
       "      <td>0.937916</td>\n",
       "      <td>0.939792</td>\n",
       "      <td>0.791484</td>\n",
       "      <td>0.856831</td>\n",
       "      <td>0.854708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27.626667</th>\n",
       "      <td>0.947415</td>\n",
       "      <td>0.845613</td>\n",
       "      <td>0.907713</td>\n",
       "      <td>0.936171</td>\n",
       "      <td>0.846106</td>\n",
       "      <td>0.959213</td>\n",
       "      <td>0.987647</td>\n",
       "      <td>0.995187</td>\n",
       "      <td>0.810169</td>\n",
       "      <td>0.587472</td>\n",
       "      <td>...</td>\n",
       "      <td>0.910162</td>\n",
       "      <td>0.803897</td>\n",
       "      <td>0.899662</td>\n",
       "      <td>0.982426</td>\n",
       "      <td>0.919437</td>\n",
       "      <td>0.928304</td>\n",
       "      <td>0.930105</td>\n",
       "      <td>0.756891</td>\n",
       "      <td>0.833297</td>\n",
       "      <td>0.832530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31.573334</th>\n",
       "      <td>0.940204</td>\n",
       "      <td>0.823702</td>\n",
       "      <td>0.897794</td>\n",
       "      <td>0.927515</td>\n",
       "      <td>0.824592</td>\n",
       "      <td>0.953748</td>\n",
       "      <td>0.986042</td>\n",
       "      <td>0.994603</td>\n",
       "      <td>0.783184</td>\n",
       "      <td>0.529048</td>\n",
       "      <td>...</td>\n",
       "      <td>0.898153</td>\n",
       "      <td>0.776638</td>\n",
       "      <td>0.885805</td>\n",
       "      <td>0.980211</td>\n",
       "      <td>0.908443</td>\n",
       "      <td>0.918692</td>\n",
       "      <td>0.920419</td>\n",
       "      <td>0.722297</td>\n",
       "      <td>0.809763</td>\n",
       "      <td>0.810352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35.520000</th>\n",
       "      <td>0.932994</td>\n",
       "      <td>0.801792</td>\n",
       "      <td>0.887876</td>\n",
       "      <td>0.918858</td>\n",
       "      <td>0.803079</td>\n",
       "      <td>0.948283</td>\n",
       "      <td>0.984437</td>\n",
       "      <td>0.994019</td>\n",
       "      <td>0.756200</td>\n",
       "      <td>0.470624</td>\n",
       "      <td>...</td>\n",
       "      <td>0.886145</td>\n",
       "      <td>0.749378</td>\n",
       "      <td>0.871949</td>\n",
       "      <td>0.977995</td>\n",
       "      <td>0.897449</td>\n",
       "      <td>0.909080</td>\n",
       "      <td>0.910732</td>\n",
       "      <td>0.687704</td>\n",
       "      <td>0.786229</td>\n",
       "      <td>0.788174</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39.466667</th>\n",
       "      <td>0.925783</td>\n",
       "      <td>0.779881</td>\n",
       "      <td>0.877957</td>\n",
       "      <td>0.910201</td>\n",
       "      <td>0.781565</td>\n",
       "      <td>0.942818</td>\n",
       "      <td>0.982831</td>\n",
       "      <td>0.993435</td>\n",
       "      <td>0.729215</td>\n",
       "      <td>0.412200</td>\n",
       "      <td>...</td>\n",
       "      <td>0.874136</td>\n",
       "      <td>0.722118</td>\n",
       "      <td>0.858092</td>\n",
       "      <td>0.975780</td>\n",
       "      <td>0.886455</td>\n",
       "      <td>0.899469</td>\n",
       "      <td>0.901046</td>\n",
       "      <td>0.653110</td>\n",
       "      <td>0.762695</td>\n",
       "      <td>0.765996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43.413334</th>\n",
       "      <td>0.902162</td>\n",
       "      <td>0.762350</td>\n",
       "      <td>0.862645</td>\n",
       "      <td>0.897250</td>\n",
       "      <td>0.754873</td>\n",
       "      <td>0.931251</td>\n",
       "      <td>0.978628</td>\n",
       "      <td>0.991024</td>\n",
       "      <td>0.685955</td>\n",
       "      <td>0.395379</td>\n",
       "      <td>...</td>\n",
       "      <td>0.863165</td>\n",
       "      <td>0.693594</td>\n",
       "      <td>0.843046</td>\n",
       "      <td>0.970013</td>\n",
       "      <td>0.866816</td>\n",
       "      <td>0.889184</td>\n",
       "      <td>0.879391</td>\n",
       "      <td>0.608127</td>\n",
       "      <td>0.742316</td>\n",
       "      <td>0.744343</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47.360001</th>\n",
       "      <td>0.878541</td>\n",
       "      <td>0.744820</td>\n",
       "      <td>0.847333</td>\n",
       "      <td>0.884298</td>\n",
       "      <td>0.728181</td>\n",
       "      <td>0.919684</td>\n",
       "      <td>0.974425</td>\n",
       "      <td>0.988613</td>\n",
       "      <td>0.642695</td>\n",
       "      <td>0.378557</td>\n",
       "      <td>...</td>\n",
       "      <td>0.852194</td>\n",
       "      <td>0.665070</td>\n",
       "      <td>0.828001</td>\n",
       "      <td>0.964245</td>\n",
       "      <td>0.847177</td>\n",
       "      <td>0.878898</td>\n",
       "      <td>0.857737</td>\n",
       "      <td>0.563145</td>\n",
       "      <td>0.721937</td>\n",
       "      <td>0.722690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51.306667</th>\n",
       "      <td>0.854920</td>\n",
       "      <td>0.727290</td>\n",
       "      <td>0.832021</td>\n",
       "      <td>0.871347</td>\n",
       "      <td>0.701489</td>\n",
       "      <td>0.908117</td>\n",
       "      <td>0.970222</td>\n",
       "      <td>0.986202</td>\n",
       "      <td>0.599434</td>\n",
       "      <td>0.361736</td>\n",
       "      <td>...</td>\n",
       "      <td>0.841223</td>\n",
       "      <td>0.636545</td>\n",
       "      <td>0.812955</td>\n",
       "      <td>0.958478</td>\n",
       "      <td>0.827538</td>\n",
       "      <td>0.868613</td>\n",
       "      <td>0.836082</td>\n",
       "      <td>0.518162</td>\n",
       "      <td>0.701558</td>\n",
       "      <td>0.701038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55.253334</th>\n",
       "      <td>0.831299</td>\n",
       "      <td>0.709760</td>\n",
       "      <td>0.816709</td>\n",
       "      <td>0.858395</td>\n",
       "      <td>0.674797</td>\n",
       "      <td>0.896550</td>\n",
       "      <td>0.966018</td>\n",
       "      <td>0.983791</td>\n",
       "      <td>0.556174</td>\n",
       "      <td>0.344915</td>\n",
       "      <td>...</td>\n",
       "      <td>0.830252</td>\n",
       "      <td>0.608021</td>\n",
       "      <td>0.797909</td>\n",
       "      <td>0.952711</td>\n",
       "      <td>0.807900</td>\n",
       "      <td>0.858328</td>\n",
       "      <td>0.814428</td>\n",
       "      <td>0.473180</td>\n",
       "      <td>0.681179</td>\n",
       "      <td>0.679385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59.200001</th>\n",
       "      <td>0.807678</td>\n",
       "      <td>0.692230</td>\n",
       "      <td>0.801397</td>\n",
       "      <td>0.845444</td>\n",
       "      <td>0.648105</td>\n",
       "      <td>0.884983</td>\n",
       "      <td>0.961815</td>\n",
       "      <td>0.981379</td>\n",
       "      <td>0.512914</td>\n",
       "      <td>0.328093</td>\n",
       "      <td>...</td>\n",
       "      <td>0.819281</td>\n",
       "      <td>0.579497</td>\n",
       "      <td>0.782863</td>\n",
       "      <td>0.946943</td>\n",
       "      <td>0.788261</td>\n",
       "      <td>0.848043</td>\n",
       "      <td>0.792773</td>\n",
       "      <td>0.428197</td>\n",
       "      <td>0.660800</td>\n",
       "      <td>0.657733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63.146667</th>\n",
       "      <td>0.784057</td>\n",
       "      <td>0.674700</td>\n",
       "      <td>0.786085</td>\n",
       "      <td>0.832492</td>\n",
       "      <td>0.621412</td>\n",
       "      <td>0.873416</td>\n",
       "      <td>0.957612</td>\n",
       "      <td>0.978968</td>\n",
       "      <td>0.469653</td>\n",
       "      <td>0.311272</td>\n",
       "      <td>...</td>\n",
       "      <td>0.808310</td>\n",
       "      <td>0.550972</td>\n",
       "      <td>0.767817</td>\n",
       "      <td>0.941176</td>\n",
       "      <td>0.768622</td>\n",
       "      <td>0.837758</td>\n",
       "      <td>0.771119</td>\n",
       "      <td>0.383215</td>\n",
       "      <td>0.640421</td>\n",
       "      <td>0.636080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67.093334</th>\n",
       "      <td>0.760436</td>\n",
       "      <td>0.657170</td>\n",
       "      <td>0.770774</td>\n",
       "      <td>0.819541</td>\n",
       "      <td>0.594720</td>\n",
       "      <td>0.861849</td>\n",
       "      <td>0.953409</td>\n",
       "      <td>0.976557</td>\n",
       "      <td>0.426393</td>\n",
       "      <td>0.294451</td>\n",
       "      <td>...</td>\n",
       "      <td>0.797339</td>\n",
       "      <td>0.522448</td>\n",
       "      <td>0.752772</td>\n",
       "      <td>0.935408</td>\n",
       "      <td>0.748983</td>\n",
       "      <td>0.827473</td>\n",
       "      <td>0.749465</td>\n",
       "      <td>0.338232</td>\n",
       "      <td>0.620041</td>\n",
       "      <td>0.614428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71.040001</th>\n",
       "      <td>0.736815</td>\n",
       "      <td>0.639639</td>\n",
       "      <td>0.755462</td>\n",
       "      <td>0.806590</td>\n",
       "      <td>0.568028</td>\n",
       "      <td>0.850281</td>\n",
       "      <td>0.949205</td>\n",
       "      <td>0.974146</td>\n",
       "      <td>0.383133</td>\n",
       "      <td>0.277629</td>\n",
       "      <td>...</td>\n",
       "      <td>0.786368</td>\n",
       "      <td>0.493924</td>\n",
       "      <td>0.737726</td>\n",
       "      <td>0.929641</td>\n",
       "      <td>0.729344</td>\n",
       "      <td>0.817188</td>\n",
       "      <td>0.727810</td>\n",
       "      <td>0.293250</td>\n",
       "      <td>0.599662</td>\n",
       "      <td>0.592775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74.986668</th>\n",
       "      <td>0.713194</td>\n",
       "      <td>0.622109</td>\n",
       "      <td>0.740150</td>\n",
       "      <td>0.793638</td>\n",
       "      <td>0.541336</td>\n",
       "      <td>0.838714</td>\n",
       "      <td>0.945002</td>\n",
       "      <td>0.971735</td>\n",
       "      <td>0.339872</td>\n",
       "      <td>0.260808</td>\n",
       "      <td>...</td>\n",
       "      <td>0.775397</td>\n",
       "      <td>0.465400</td>\n",
       "      <td>0.722680</td>\n",
       "      <td>0.923874</td>\n",
       "      <td>0.709706</td>\n",
       "      <td>0.806903</td>\n",
       "      <td>0.706156</td>\n",
       "      <td>0.248267</td>\n",
       "      <td>0.579283</td>\n",
       "      <td>0.571123</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>20 rows × 381 columns</p>\n",
       "</div>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 6
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Build the evaluator.\n",
    "Some arguments includes:\n",
    "1. Training set information (`durations_train` and `events_train`) is optional if you don't need Brier score, integrated Brier score, and L1-margin loss.\n",
    "2. `predicted_time_method` argument indicates whether using median or mean survival time as the predicted time."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "eval = PycoxEvaluator(surv, durations_test, events_test, durations_train, events_train, predict_time_method=\"Median\")"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:55.132981Z",
     "start_time": "2024-04-28T21:03:55.130400Z"
    }
   },
   "outputs": [],
   "execution_count": 7
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Make some example plots\n",
    "Plot the survival curves for subject 1, 10 and 100."
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "eval.plot_survival_curves([1, 10, 100])"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:57.043866Z",
     "start_time": "2024-04-28T21:03:56.921788Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<Figure size 640x480 with 1 Axes>,\n",
       " <Axes: xlabel='Time', ylabel='Survival probability'>)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 8
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Calculate the concordance index.\n",
    "`ties` argument indicates the way ties should be handled.\n",
    "`ties=\"None\"` will throw out all ties in true survival time and all ties in predict survival times (risk scores).\n",
    "`ties=\"Time\"` includes ties in true survival time but removes ties in predict survival times (risk scores).\n",
    "`ties=\"Risk\"` includes ties in predict survival times (risk scores) but not in true survival time.\n",
    "`ties=\"All\"` includes all ties."
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   }
  },
  {
   "cell_type": "code",
   "source": [
    "cindex, concordant_pairs, total_pairs = eval.concordance(ties=\"None\")\n",
    "print(\"Concordance index is {}, meaning that the model can correctly order {} pairs among {} comparable pairs \"\n",
    "      \"in the testing set.\".format(cindex, concordant_pairs, total_pairs))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:03:58.685767Z",
     "start_time": "2024-04-28T21:03:58.673523Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Concordance index is 0.6397049472211624, meaning that the model can correctly order 30180.0 pairs among 47178.0 comparable pairs in the testing set.\n"
     ]
    }
   ],
   "execution_count": 9
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Calculate the Brier score\n",
    "Target time is at time 175."
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "bs = eval.brier_score(target_time=175)\n",
    "print(\"Brier score at time 175 is {}.\".format(bs))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:00.768419Z",
     "start_time": "2024-04-28T21:04:00.721956Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brier score at time 175 is 0.2150260862030445.\n"
     ]
    }
   ],
   "execution_count": 10
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Calculate the integrated Brier score (IBS)\n",
    "- You can specify the number of points `num_points` to integral. If it is not specify, the default setting will use the censored time points obtained from testing set.\n",
    "- You can also set the `IPCW_weighted` as True or False to calculate the IPCW-weighted IBS or non-weighted IBS. The default setting is True."
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "ibs = eval.integrated_brier_score(num_points=None, IPCW_weighted=False, draw_figure=True)\n",
    "print(\"Integrated Brier score is {}.\".format(ibs))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:03.778766Z",
     "start_time": "2024-04-28T21:04:03.586247Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Integrated Brier score is 0.10367363212850687.\n"
     ]
    }
   ],
   "execution_count": 11
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Calculate the MAE-Hinge loss\n",
    "You can set `method` as `Hinge`, `Uncensored`, `Margin`, `IPCW-v1`, `IPCW-v2`, or `Pseudo_obs` for different MAE calculation."
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "mae_score = eval.mae(method='Hinge')\n",
    "print(\"MAE-hinge loss is {}.\".format(mae_score))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:08.115158Z",
     "start_time": "2024-04-28T21:04:08.112196Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MAE-hinge loss is 61.36490119533371.\n"
     ]
    }
   ],
   "execution_count": 12
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Calculate the 1-Calibration\n",
    "Run one calibration test at 50% time of the testing set."
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "t_time = round(np.percentile(durations_test, 50))\n",
    "p_value, ob, exp = eval.one_calibration(target_time=t_time, method=\"DN\")\n",
    "if p_value >= 0.05:\n",
    "    print(\"The model is one-calibrated at time {}.\".format(t_time))\n",
    "else:\n",
    "    print(\"The model is not one-calibrated at time {}.\".format(t_time))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:09.609541Z",
     "start_time": "2024-04-28T21:04:09.567533Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The model is one-calibrated at time 113.\n"
     ]
    }
   ],
   "execution_count": 13
  },
  {
   "cell_type": "markdown",
   "source": [
    "###  Plot the histogram for 1-Calibration.\n",
    "As presented in [Effective Ways to Build and Evaluate Individual Survival Distributions](https://jmlr.csail.mit.edu/papers/v21/18-772.html)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "# Reverse the bins to let it start from small to large observations\n",
    "ob.reverse()\n",
    "exp.reverse()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:11.064082Z",
     "start_time": "2024-04-28T21:04:11.062053Z"
    }
   },
   "outputs": [],
   "execution_count": 14
  },
  {
   "cell_type": "code",
   "source": [
    "bins = np.arange(10)\n",
    "plt.bar(bins - 0.15, ob, width=0.3, label=\"Observed\")\n",
    "plt.bar(bins + 0.15, exp, width=0.3, label=\"Expected\")\n",
    "\n",
    "plt.xlabel('Bin')\n",
    "plt.ylabel('Survival Probability')\n",
    "plt.text(0.5, 0.5, r'$p-value = {:.3f}$'.format(p_value), verticalalignment='top', horizontalalignment='left', fontsize=12, color='Black')\n",
    "plt.legend()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:11.789852Z",
     "start_time": "2024-04-28T21:04:11.672698Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 15
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Calculate the distribution calibration (D-Calibration)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "p_value_2, bin_statistics = eval.d_calibration()\n",
    "if p_value_2 >= 0.05:\n",
    "    print(\"The model is d-calibrated!\")\n",
    "else:\n",
    "    print(\"The model is not d-calibrated!\")"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:13.202692Z",
     "start_time": "2024-04-28T21:04:13.174515Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The model is d-calibrated!\n"
     ]
    }
   ],
   "execution_count": 16
  },
  {
   "cell_type": "code",
   "source": [
    "# An ideal model should have bin statistics equally distributed. That means 0.1 for each value since we are using 10 bins.\n",
    "bin_statistics /= bin_statistics.sum()\n",
    "bin_statistics"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:13.737678Z",
     "start_time": "2024-04-28T21:04:13.734679Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.09007839, 0.10882139, 0.0850883 , 0.0803543 , 0.10668296,\n",
       "       0.0890336 , 0.09686899, 0.09676666, 0.11162852, 0.13467688])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 17
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Plot the sideways histogram of D-Calibration\n",
    "As presented in [Effective Ways to Build and Evaluate Individual Survival Distributions](https://jmlr.csail.mit.edu/papers/v21/18-772.html)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "bars = ('[0.0, 0.1)', '[0.1, 0.2)', '[0.2, 0.3)', '[0.3, 0.4)', '[0.4, 0.5)',\n",
    "        '[0.5, 0.6)', '[0.6, 0.7)', '[0.7, 0.8)', '[0.8, 0.9)', '[0.9, 1.0]')\n",
    "y_pos = np.arange(len(bars))\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "ax.barh(y_pos, bin_statistics)\n",
    "ax.plot(np.ones(len(bars)+2) * 0.1, np.arange(-1, len(bars)+1),  '--', color=\"black\")\n",
    "ax.invert_yaxis()  # labels read top-to-bottom\n",
    "ax.set_xlabel('Proportion in Bin')\n",
    "ax.set_ylabel('Bins')\n",
    "ax.text(0.14, 0.5, r'$p-value = {:.2f}$'.format(p_value_2), verticalalignment='top', horizontalalignment='right', fontsize=10, color='Black')\n",
    "plt.ylim([-0.8, 9.8])\n",
    "plt.yticks(y_pos, bars)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:14.870327Z",
     "start_time": "2024-04-28T21:04:14.783862Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 18
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Plot the Q-Q plot of D-Calibration\n",
    "As presented in [Countdown Regression: Sharp and Calibrated Survival Predictions](https://proceedings.mlr.press/v115/avati20a.html) and [Inverse-Weighted Survival Games](https://arxiv.org/abs/2111.08175)"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "source": [
    "default = np.linspace(0, 1, 11)\n",
    "plt.plot(default, default, linestyle='dashed', label=\"x=y\")\n",
    "\n",
    "fraction = np.concatenate((np.array([0]), bin_statistics))\n",
    "plt.plot(default, fraction.cumsum(), 'o-', label=\"LogisticHazard\")\n",
    "\n",
    "plt.xlabel(\"Cumulative Risk\")\n",
    "plt.ylabel(\"Observed Cumulative Fraction\")\n",
    "plt.legend()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "ExecuteTime": {
     "end_time": "2024-04-28T21:04:16.556972Z",
     "start_time": "2024-04-28T21:04:16.476688Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 19
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python3",
   "language": "python",
   "display_name": "Python 3 (ipykernel)"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}