{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a578f2ab",
   "metadata": {},
   "source": [
    "# Virny Demo for the ACS Income Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2251a923",
   "metadata": {},
   "source": [
    "In this example, we are going to apply [Virny](https://github.com/DataResponsibly/Virny) to conduct a deep performance profiling for 5 models trained on the [ACS Income dataset](https://github.com/zykls/folktables). This demonstration will show how to create input arguments for Virny, how to compute overall and disparity metrics with a metric computation interface, and how to build static and interactive visualizations based on the calculated metrics. \n",
    "\n",
    "The structure of this notebook is the following:\n",
    "* **Step 1**: Create a _config yaml_ for metric computation.\n",
    "* **Step 2**: Preprocess a dataset and construct a _BaseFlowDataset_ object.\n",
    "* **Step 3**: Tune models and create a _models config_.\n",
    "* **Step 4**: Run a metric computation interface from Virny.\n",
    "* **Step 5**: Compose disparity metrics using _Metric Composer_.\n",
    "* **Step 6**: Create static visualizations using _Metric Static Visualizer_.\n",
    "* **Step 7**: Build an interactive web app using _Metric Interactive Visualizer_.\n",
    "\n",
    "To view other use case examples that demonstrate all Virny capabilities, you can go to [this webpage](https://dataresponsibly.github.io/Virny/examples/Multiple_Models_Interface_Use_Case/)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "606df34d",
   "metadata": {},
   "source": [
    "## Install necessary packages and import dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "outputs": [],
   "source": [
    "# Install Virny using pypi. The library supports Python 3.8 and 3.9.\n",
    "!pip install virny==0.5.0\n",
    "!pip install xgboost==1.7.2"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:05:54.988254Z",
     "start_time": "2024-06-02T14:05:54.960443Z"
    }
   },
   "id": "668af6f1babe4564"
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import os\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "os.environ[\"PYTHONWARNINGS\"] = \"ignore\""
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:20.738385Z",
     "start_time": "2024-06-02T14:01:20.333024Z"
    }
   },
   "id": "635f5cae205d3b4f"
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "7a9241de",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-06-02T14:13:58.259297Z",
     "start_time": "2024-06-02T14:13:58.202764Z"
    }
   },
   "outputs": [],
   "source": [
    "from pprint import pprint\n",
    "from datetime import datetime, timezone\n",
    "\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "from virny.utils.custom_initializers import create_config_obj, read_model_metric_dfs\n",
    "from virny.user_interfaces.multiple_models_api import compute_metrics_with_config\n",
    "from virny.preprocessing.basic_preprocessing import preprocess_dataset\n",
    "from virny.custom_classes.metrics_interactive_visualizer import MetricsInteractiveVisualizer\n",
    "from virny.custom_classes.metrics_visualizer import MetricsVisualizer\n",
    "from virny.custom_classes.metrics_composer import MetricsComposer\n",
    "from virny.utils.model_tuning_utils import tune_ML_models"
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 1**: Create a _config yaml_ for metrics computation."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "154ef968c9937432"
  },
  {
   "cell_type": "markdown",
   "source": [
    "First, we need to create a _config yaml_, which includes the following parameters for metrics computation:\n",
    "\n",
    "* **dataset_name**: str, a name of your dataset; it will be used to name files with metrics.\n",
    "\n",
    "* **bootstrap_fraction**: float, the fraction from a train set in the range [0.0 - 1.0] to fit models in bootstrap (usually more than 0.5).\n",
    "\n",
    "* **random_state**: int, a seed to control the randomness of the whole model evaluation pipeline.\n",
    "\n",
    "* **n_estimators**: int, the number of estimators for bootstrap to compute stability and uncertainty metrics.\n",
    "\n",
    "* **sensitive_attributes_dct**: dict, a dictionary where keys are sensitive attribute names (including intersectional attributes), and values are disadvantaged values for these attributes. Intersectional attributes must include '&' between sensitive attributes. You do not need to specify disadvantaged values for intersectional groups since they will be derived from disadvantaged values in sensitive_attributes_dct for each separate sensitive attribute in this intersectional pair."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "8ae47aed99d5ba8d"
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "outputs": [],
   "source": [
    "config_yaml_path = os.path.join('.', 'experiment_config.yaml')\n",
    "config_yaml_content = \"\"\"\n",
    "dataset_name: ACS_Income\n",
    "bootstrap_fraction: 0.8\n",
    "random_state: 42\n",
    "n_estimators: 20  # Better to input the higher number of estimators than 100; this is only for this demonstration\n",
    "sensitive_attributes_dct: {'SEX': '2', 'RAC1P': ['2', '3', '4', '5', '6', '7', '8', '9'], 'SEX & RAC1P': None}\n",
    "\"\"\"\n",
    "\n",
    "with open(config_yaml_path, 'w', encoding='utf-8') as f:\n",
    "    f.write(config_yaml_content)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:22.872204Z",
     "start_time": "2024-06-02T14:01:22.844817Z"
    }
   },
   "id": "bbd56e951b5fd132"
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "outputs": [],
   "source": [
    "config = create_config_obj(config_yaml_path=config_yaml_path)\n",
    "SAVE_RESULTS_DIR_PATH = os.path.join('.', 'results', f'{config.dataset_name}_Metrics_{datetime.now(timezone.utc).strftime(\"%Y%m%d__%H%M%S\")}')"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:22.902168Z",
     "start_time": "2024-06-02T14:01:22.873186Z"
    }
   },
   "id": "b2fd18d315063a47"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 2**: Preprocess a dataset and construct a _BaseFlowDataset_ object."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "8c633b771e77945a"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Second, we need to import a dataset and preprocess it. In this demonstration, we are working with the ACS Income dataset, but you can use [any data loader provided by Virny](https://dataresponsibly.github.io/Virny/api/overview/#datasets) or create yours as described in [this example](https://dataresponsibly.github.io/Virny/examples/Multiple_Models_Interface_Use_Case/#preprocess-the-dataset-and-create-a-baseflowdataset-class).\n",
    "\n",
    "Note that if you get an error importing the ACS Income dataset, this means that it is not available for your geographic location. In this case, you can use VPN to overcome this issue."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "5f93793c51964f51"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Import the dataset."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "38fba7f2330b0ae9"
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "outputs": [
    {
     "data": {
      "text/plain": "  SCHL COW MAR  OCCP POBP RELP SEX RAC1P  AGEP  WKHP  PINCP\n0   23   7   3   230   36    0   1     1    55  55.0      1\n1   16   1   5  4110   13    2   2     1    20  35.0      0\n2   16   4   3  4130   51    0   2     1    59  30.0      0\n3   18   4   1  4020   13    0   1     2    43  40.0      0\n4   14   1   1  8300   20    1   2     2    33  20.0      0",
      "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>SCHL</th>\n      <th>COW</th>\n      <th>MAR</th>\n      <th>OCCP</th>\n      <th>POBP</th>\n      <th>RELP</th>\n      <th>SEX</th>\n      <th>RAC1P</th>\n      <th>AGEP</th>\n      <th>WKHP</th>\n      <th>PINCP</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>23</td>\n      <td>7</td>\n      <td>3</td>\n      <td>230</td>\n      <td>36</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>55</td>\n      <td>55.0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>16</td>\n      <td>1</td>\n      <td>5</td>\n      <td>4110</td>\n      <td>13</td>\n      <td>2</td>\n      <td>2</td>\n      <td>1</td>\n      <td>20</td>\n      <td>35.0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>16</td>\n      <td>4</td>\n      <td>3</td>\n      <td>4130</td>\n      <td>51</td>\n      <td>0</td>\n      <td>2</td>\n      <td>1</td>\n      <td>59</td>\n      <td>30.0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>18</td>\n      <td>4</td>\n      <td>1</td>\n      <td>4020</td>\n      <td>13</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2</td>\n      <td>43</td>\n      <td>40.0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>14</td>\n      <td>1</td>\n      <td>1</td>\n      <td>8300</td>\n      <td>20</td>\n      <td>1</td>\n      <td>2</td>\n      <td>2</td>\n      <td>33</td>\n      <td>20.0</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from virny.datasets import ACSIncomeDataset\n",
    "\n",
    "data_loader = ACSIncomeDataset(state=['GA'], year=2018, with_nulls=False,\n",
    "                               subsample_size=15_000, subsample_seed=42)\n",
    "data_loader.full_df.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:24.193309Z",
     "start_time": "2024-06-02T14:01:22.898511Z"
    }
   },
   "id": "7bd69e89d33f3e08"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Define preprocessing steps and initialize a column transformer."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "987fbf1fe5834390"
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [],
   "source": [
    "column_transformer = ColumnTransformer(transformers=[\n",
    "    ('categorical_features', OneHotEncoder(handle_unknown='ignore', sparse=False), data_loader.categorical_columns),\n",
    "    ('numerical_features', StandardScaler(), data_loader.numerical_columns),\n",
    "])"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:24.204251Z",
     "start_time": "2024-06-02T14:01:24.162607Z"
    }
   },
   "id": "158baf3f0853864e"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Construct a BaseFlowDataset object."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "6eaa9c791d08d504"
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "outputs": [],
   "source": [
    "DATASET_SPLIT_SEED = 42\n",
    "MODELS_TUNING_SEED = 42\n",
    "TEST_SET_FRACTION = 0.2\n",
    "\n",
    "base_flow_dataset = preprocess_dataset(data_loader=data_loader,\n",
    "                                       column_transformer=column_transformer,\n",
    "                                       sensitive_attributes_dct=config.sensitive_attributes_dct,\n",
    "                                       test_set_fraction=TEST_SET_FRACTION,\n",
    "                                       dataset_split_seed=DATASET_SPLIT_SEED)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:24.324734Z",
     "start_time": "2024-06-02T14:01:24.188842Z"
    }
   },
   "id": "213b893a804b4bd6"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 3**: Tune models and create a _models config_."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "483da592b29ba485"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Next, we need to construct a _models config_ that includes initialized models you want to profile with Virny. For that, the models should be tuned using the _tune_ML_models()_ function from Virny or in any other convenient way.\n",
    "\n",
    "Note that the model name convention for the _models config_ is '<MODEL_TYPE>' or '<MODEL_TYPE>__<ANY_SUFFIX>' (with two underscores), for example, 'DecisionTreeClassifier' or 'DecisionTreeClassifier__alpha=0.6'. This convention is needed to create appropriate bar charts for model selection in the interactive web app."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "990330e9d7da5e70"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Define models and hyper-parameters to tune using GridSearchCV from sklearn."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "660af59469317200"
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "outputs": [],
   "source": [
    "models_params_for_tuning = {\n",
    "    'DecisionTreeClassifier': {\n",
    "        'model': DecisionTreeClassifier(random_state=MODELS_TUNING_SEED),\n",
    "        'params': {\n",
    "            \"max_depth\": [20, 30],\n",
    "            \"min_samples_split\" : [0.1],\n",
    "            \"max_features\": ['sqrt'],\n",
    "            \"criterion\": [\"gini\", \"entropy\"]\n",
    "        }\n",
    "    },\n",
    "    'LogisticRegression': {\n",
    "        'model': LogisticRegression(random_state=MODELS_TUNING_SEED),\n",
    "        'params': {\n",
    "            'penalty': ['l2'],\n",
    "            'C' : [0.0001, 0.1, 1, 100],\n",
    "            'solver': ['newton-cg', 'lbfgs'],\n",
    "            'max_iter': [250],\n",
    "        }\n",
    "    },\n",
    "    'KNeighborsClassifier': {\n",
    "        'model': KNeighborsClassifier(),\n",
    "        'params': {\n",
    "            'weights' : ['uniform'],\n",
    "            'algorithm' : ['auto'],\n",
    "            'n_neighbors' : [3, 4, 5],\n",
    "        }\n",
    "    },\n",
    "    'RandomForestClassifier': {\n",
    "        'model': RandomForestClassifier(random_state=MODELS_TUNING_SEED),\n",
    "        'params': {\n",
    "            \"max_depth\": [10],\n",
    "            \"min_samples_leaf\": [1],\n",
    "            \"n_estimators\": [20, 50],\n",
    "            \"max_features\": [0.6]\n",
    "        }\n",
    "    },\n",
    "    'XGBClassifier': {\n",
    "        'model': XGBClassifier(random_state=MODELS_TUNING_SEED, verbosity=0),\n",
    "        'params': {\n",
    "            'learning_rate': [0.1],\n",
    "            'n_estimators': [50],\n",
    "            'max_depth': [5, 7],\n",
    "            'lambda':  [10, 100]\n",
    "        }\n",
    "    }\n",
    "}"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:24.599986Z",
     "start_time": "2024-06-02T14:01:24.290509Z"
    }
   },
   "id": "e1badaf3e65e97e"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Tune models using the _tune_ML_models()_ function from Virny and create the _models config_."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "5a267ffdf55fa180"
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024/06/02, 17:01:24: Tuning DecisionTreeClassifier...\n",
      "2024/06/02, 17:01:26: Tuning for DecisionTreeClassifier is finished [F1 score = 0.689443059507747, Accuracy = 0.7349166666666668]\n",
      "\n",
      "2024/06/02, 17:01:26: Tuning LogisticRegression...\n",
      "2024/06/02, 17:01:35: Tuning for LogisticRegression is finished [F1 score = 0.7929754140358044, Accuracy = 0.8145000000000001]\n",
      "\n",
      "2024/06/02, 17:01:35: Tuning KNeighborsClassifier...\n",
      "2024/06/02, 17:01:37: Tuning for KNeighborsClassifier is finished [F1 score = 0.7499760379663827, Accuracy = 0.7724166666666666]\n",
      "\n",
      "2024/06/02, 17:01:37: Tuning RandomForestClassifier...\n",
      "2024/06/02, 17:01:40: Tuning for RandomForestClassifier is finished [F1 score = 0.7663327842026387, Accuracy = 0.7935833333333333]\n",
      "\n",
      "2024/06/02, 17:01:40: Tuning XGBClassifier...\n",
      "2024/06/02, 17:01:53: Tuning for XGBClassifier is finished [F1 score = 0.7754257292923424, Accuracy = 0.8003333333333332]\n"
     ]
    },
    {
     "data": {
      "text/plain": "  Dataset_Name              Model_Name  F1_Score  Accuracy_Score  \\\n0   ACS_Income  DecisionTreeClassifier  0.689443        0.734917   \n1   ACS_Income      LogisticRegression  0.792975        0.814500   \n2   ACS_Income    KNeighborsClassifier  0.749976        0.772417   \n3   ACS_Income  RandomForestClassifier  0.766333        0.793583   \n4   ACS_Income           XGBClassifier  0.775426        0.800333   \n\n                                   Model_Best_Params  \n0  {'criterion': 'entropy', 'max_depth': 20, 'max...  \n1  {'C': 1, 'max_iter': 250, 'penalty': 'l2', 'so...  \n2  {'algorithm': 'auto', 'n_neighbors': 5, 'weigh...  \n3  {'max_depth': 10, 'max_features': 0.6, 'min_sa...  \n4  {'lambda': 10, 'learning_rate': 0.1, 'max_dept...  ",
      "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>Dataset_Name</th>\n      <th>Model_Name</th>\n      <th>F1_Score</th>\n      <th>Accuracy_Score</th>\n      <th>Model_Best_Params</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>ACS_Income</td>\n      <td>DecisionTreeClassifier</td>\n      <td>0.689443</td>\n      <td>0.734917</td>\n      <td>{'criterion': 'entropy', 'max_depth': 20, 'max...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>ACS_Income</td>\n      <td>LogisticRegression</td>\n      <td>0.792975</td>\n      <td>0.814500</td>\n      <td>{'C': 1, 'max_iter': 250, 'penalty': 'l2', 'so...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>ACS_Income</td>\n      <td>KNeighborsClassifier</td>\n      <td>0.749976</td>\n      <td>0.772417</td>\n      <td>{'algorithm': 'auto', 'n_neighbors': 5, 'weigh...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>ACS_Income</td>\n      <td>RandomForestClassifier</td>\n      <td>0.766333</td>\n      <td>0.793583</td>\n      <td>{'max_depth': 10, 'max_features': 0.6, 'min_sa...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>ACS_Income</td>\n      <td>XGBClassifier</td>\n      <td>0.775426</td>\n      <td>0.800333</td>\n      <td>{'lambda': 10, 'learning_rate': 0.1, 'max_dept...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tuned_params_df, models_config = tune_ML_models(models_params_for_tuning, base_flow_dataset, dataset_name='ACS_Income', n_folds=3)\n",
    "tuned_params_df"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:53.850562Z",
     "start_time": "2024-06-02T14:01:24.314934Z"
    }
   },
   "id": "a7577ac3a3fb389d"
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'DecisionTreeClassifier': DecisionTreeClassifier(criterion='entropy', max_depth=20, max_features='sqrt',\n",
      "                       min_samples_split=0.1, random_state=42),\n",
      " 'KNeighborsClassifier': KNeighborsClassifier(),\n",
      " 'LogisticRegression': LogisticRegression(C=1, max_iter=250, random_state=42, solver='newton-cg'),\n",
      " 'RandomForestClassifier': RandomForestClassifier(max_depth=10, max_features=0.6, n_estimators=20,\n",
      "                       random_state=42),\n",
      " 'XGBClassifier': XGBClassifier(base_score=None, booster=None, callbacks=None,\n",
      "              colsample_bylevel=None, colsample_bynode=None,\n",
      "              colsample_bytree=None, early_stopping_rounds=None,\n",
      "              enable_categorical=False, eval_metric=None, feature_types=None,\n",
      "              gamma=None, gpu_id=None, grow_policy=None, importance_type=None,\n",
      "              interaction_constraints=None, lambda=10, learning_rate=0.1,\n",
      "              max_bin=None, max_cat_threshold=None, max_cat_to_onehot=None,\n",
      "              max_delta_step=None, max_depth=7, max_leaves=None,\n",
      "              min_child_weight=None, missing=nan, monotone_constraints=None,\n",
      "              n_estimators=50, n_jobs=None, num_parallel_tree=None,\n",
      "              predictor=None, ...)}\n"
     ]
    }
   ],
   "source": [
    "pprint(models_config)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:01:53.957346Z",
     "start_time": "2024-06-02T14:01:53.847255Z"
    }
   },
   "id": "9bcb83f9f76ab5b"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 4**: Run a metric computation interface from Virny."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "bf80fdad007c7456"
  },
  {
   "cell_type": "markdown",
   "source": [
    "After that we need to input the _BaseFlowDataset_ object, _models config_, and _config yaml_ to a metric computation interface and execute it. The interface uses subgroup analyzers to compute different sets of metrics for each privileged and disadvantaged group. As for now, our library supports **Subgroup Variance Analyzer** and **Subgroup Error Analyzer**, but it is easily extensible to any other analyzers. When the variance and error analyzers complete metric computation, their metrics are combined, returned in a matrix format, and stored in a file if defined."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "52622dd736aa6046"
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "outputs": [
    {
     "data": {
      "text/plain": "Analyze multiple models:   0%|          | 0/5 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "519052a7b2df4fc2be8a7ff0592f3e22"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Classifiers testing by bootstrap:   0%|          | 0/20 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "c856bcdd21524e5cb330d3473be25b2f"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Classifiers testing by bootstrap:   0%|          | 0/20 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "b2aa86f35dfd4b43932a0c873a837595"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Classifiers testing by bootstrap:   0%|          | 0/20 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "077bc8a511aa4497962e65bfe49b20b6"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Classifiers testing by bootstrap:   0%|          | 0/20 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "04ef0b48338b402c849e3a3339965462"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Classifiers testing by bootstrap:   0%|          | 0/20 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "14ed8a968a0f4d5aa7ca480e6ad4a214"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "metrics_dct = compute_metrics_with_config(base_flow_dataset, config, models_config, SAVE_RESULTS_DIR_PATH, notebook_logs_stdout=True)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:24.852725Z",
     "start_time": "2024-06-02T14:01:53.895989Z"
    }
   },
   "id": "5846f004634b1c5a"
  },
  {
   "cell_type": "markdown",
   "source": [
    "View the computed metrics for one model."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "e085d612eca8e69f"
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "outputs": [
    {
     "data": {
      "text/plain": "                   Metric      overall     SEX_priv      SEX_dis   RAC1P_priv  \\\n0        Statistical_Bias     0.358675     0.380756     0.335571     0.368035   \n1         Mean_Prediction     0.649340     0.605822     0.694878     0.622922   \n2     Overall_Uncertainty     0.840542     0.865386     0.814545     0.857158   \n3   Aleatoric_Uncertainty     0.754194     0.779763     0.727440     0.768220   \n4                     Std     0.147406     0.149878     0.144821     0.151499   \n5                     IQR     0.203188     0.204853     0.201446     0.212635   \n6   Epistemic_Uncertainty     0.086348     0.085624     0.087105     0.088938   \n7         Label_Stability     0.647967     0.600652     0.697476     0.599409   \n8                  Jitter     0.246075     0.279682     0.210910     0.275395   \n9                     TPR     0.472816     0.545741     0.356061     0.546392   \n10                    TNR     0.945685     0.905556     0.979439     0.925040   \n11                    PPV     0.819865     0.802784     0.865031     0.818533   \n12                    FNR     0.527184     0.454259     0.643939     0.453608   \n13                    FPR     0.054315     0.094444     0.020561     0.074960   \n14               Accuracy     0.783333     0.756845     0.811050     0.780296   \n15                     F1     0.599754     0.649765     0.504472     0.655332   \n16         Selection-Rate     0.198000     0.280965     0.111187     0.255172   \n17            Sample_Size  3000.000000  1534.000000  1466.000000  2030.000000   \n\n     RAC1P_dis  SEX&RAC1P_priv  SEX&RAC1P_dis              Model_Name  \\\n0     0.339087        0.366738       0.319588  DecisionTreeClassifier   \n1     0.704629        0.633631       0.725500  DecisionTreeClassifier   \n2     0.805768        0.852220       0.783926  DecisionTreeClassifier   \n3     0.724842        0.764880       0.702390  DecisionTreeClassifier   \n4     0.138842        0.149567       0.136930  DecisionTreeClassifier   \n5     0.183419        0.207214       0.183673  DecisionTreeClassifier   \n6     0.080926        0.087340       0.081536  DecisionTreeClassifier   \n7     0.749588        0.623000       0.769006  DecisionTreeClassifier   \n8     0.184715        0.262096       0.168411  DecisionTreeClassifier   \n9     0.248031        0.509290       0.182609  DecisionTreeClassifier   \n10    0.981844        0.934478       0.989950  DecisionTreeClassifier   \n11    0.828947        0.818981       0.840000  DecisionTreeClassifier   \n12    0.751969        0.490710       0.817391  DecisionTreeClassifier   \n13    0.018156        0.065522       0.010050  DecisionTreeClassifier   \n14    0.789691        0.778046       0.808967  DecisionTreeClassifier   \n15    0.381818        0.628032       0.300000  DecisionTreeClassifier   \n16    0.078351        0.228790       0.048733  DecisionTreeClassifier   \n17  970.000000     2487.000000     513.000000  DecisionTreeClassifier   \n\n                                         Model_Params  Virny_Random_State  \\\n0   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n1   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n2   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n3   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n4   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n5   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n6   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n7   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n8   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n9   {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n10  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n11  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n12  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n13  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n14  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n15  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n16  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n17  {'ccp_alpha': 0.0, 'class_weight': None, 'crit...                  42   \n\n    Runtime_in_Mins  \n0          0.047825  \n1          0.047825  \n2          0.047825  \n3          0.047825  \n4          0.047825  \n5          0.047825  \n6          0.047825  \n7          0.047825  \n8          0.047825  \n9          0.047825  \n10         0.047825  \n11         0.047825  \n12         0.047825  \n13         0.047825  \n14         0.047825  \n15         0.047825  \n16         0.047825  \n17         0.047825  ",
      "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>Metric</th>\n      <th>overall</th>\n      <th>SEX_priv</th>\n      <th>SEX_dis</th>\n      <th>RAC1P_priv</th>\n      <th>RAC1P_dis</th>\n      <th>SEX&amp;RAC1P_priv</th>\n      <th>SEX&amp;RAC1P_dis</th>\n      <th>Model_Name</th>\n      <th>Model_Params</th>\n      <th>Virny_Random_State</th>\n      <th>Runtime_in_Mins</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Statistical_Bias</td>\n      <td>0.358675</td>\n      <td>0.380756</td>\n      <td>0.335571</td>\n      <td>0.368035</td>\n      <td>0.339087</td>\n      <td>0.366738</td>\n      <td>0.319588</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Mean_Prediction</td>\n      <td>0.649340</td>\n      <td>0.605822</td>\n      <td>0.694878</td>\n      <td>0.622922</td>\n      <td>0.704629</td>\n      <td>0.633631</td>\n      <td>0.725500</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Overall_Uncertainty</td>\n      <td>0.840542</td>\n      <td>0.865386</td>\n      <td>0.814545</td>\n      <td>0.857158</td>\n      <td>0.805768</td>\n      <td>0.852220</td>\n      <td>0.783926</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Aleatoric_Uncertainty</td>\n      <td>0.754194</td>\n      <td>0.779763</td>\n      <td>0.727440</td>\n      <td>0.768220</td>\n      <td>0.724842</td>\n      <td>0.764880</td>\n      <td>0.702390</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Std</td>\n      <td>0.147406</td>\n      <td>0.149878</td>\n      <td>0.144821</td>\n      <td>0.151499</td>\n      <td>0.138842</td>\n      <td>0.149567</td>\n      <td>0.136930</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>IQR</td>\n      <td>0.203188</td>\n      <td>0.204853</td>\n      <td>0.201446</td>\n      <td>0.212635</td>\n      <td>0.183419</td>\n      <td>0.207214</td>\n      <td>0.183673</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>Epistemic_Uncertainty</td>\n      <td>0.086348</td>\n      <td>0.085624</td>\n      <td>0.087105</td>\n      <td>0.088938</td>\n      <td>0.080926</td>\n      <td>0.087340</td>\n      <td>0.081536</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>Label_Stability</td>\n      <td>0.647967</td>\n      <td>0.600652</td>\n      <td>0.697476</td>\n      <td>0.599409</td>\n      <td>0.749588</td>\n      <td>0.623000</td>\n      <td>0.769006</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>Jitter</td>\n      <td>0.246075</td>\n      <td>0.279682</td>\n      <td>0.210910</td>\n      <td>0.275395</td>\n      <td>0.184715</td>\n      <td>0.262096</td>\n      <td>0.168411</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>TPR</td>\n      <td>0.472816</td>\n      <td>0.545741</td>\n      <td>0.356061</td>\n      <td>0.546392</td>\n      <td>0.248031</td>\n      <td>0.509290</td>\n      <td>0.182609</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>TNR</td>\n      <td>0.945685</td>\n      <td>0.905556</td>\n      <td>0.979439</td>\n      <td>0.925040</td>\n      <td>0.981844</td>\n      <td>0.934478</td>\n      <td>0.989950</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>PPV</td>\n      <td>0.819865</td>\n      <td>0.802784</td>\n      <td>0.865031</td>\n      <td>0.818533</td>\n      <td>0.828947</td>\n      <td>0.818981</td>\n      <td>0.840000</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>FNR</td>\n      <td>0.527184</td>\n      <td>0.454259</td>\n      <td>0.643939</td>\n      <td>0.453608</td>\n      <td>0.751969</td>\n      <td>0.490710</td>\n      <td>0.817391</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>FPR</td>\n      <td>0.054315</td>\n      <td>0.094444</td>\n      <td>0.020561</td>\n      <td>0.074960</td>\n      <td>0.018156</td>\n      <td>0.065522</td>\n      <td>0.010050</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>14</th>\n      <td>Accuracy</td>\n      <td>0.783333</td>\n      <td>0.756845</td>\n      <td>0.811050</td>\n      <td>0.780296</td>\n      <td>0.789691</td>\n      <td>0.778046</td>\n      <td>0.808967</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>15</th>\n      <td>F1</td>\n      <td>0.599754</td>\n      <td>0.649765</td>\n      <td>0.504472</td>\n      <td>0.655332</td>\n      <td>0.381818</td>\n      <td>0.628032</td>\n      <td>0.300000</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>16</th>\n      <td>Selection-Rate</td>\n      <td>0.198000</td>\n      <td>0.280965</td>\n      <td>0.111187</td>\n      <td>0.255172</td>\n      <td>0.078351</td>\n      <td>0.228790</td>\n      <td>0.048733</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n    <tr>\n      <th>17</th>\n      <td>Sample_Size</td>\n      <td>3000.000000</td>\n      <td>1534.000000</td>\n      <td>1466.000000</td>\n      <td>2030.000000</td>\n      <td>970.000000</td>\n      <td>2487.000000</td>\n      <td>513.000000</td>\n      <td>DecisionTreeClassifier</td>\n      <td>{'ccp_alpha': 0.0, 'class_weight': None, 'crit...</td>\n      <td>42</td>\n      <td>0.047825</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_model_metrics_df = metrics_dct[list(models_config.keys())[0]]\n",
    "sample_model_metrics_df.head(20)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:24.889615Z",
     "start_time": "2024-06-02T14:03:24.852962Z"
    }
   },
   "id": "4f5944783bc37d08"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 5**: Compose disparity metrics using _Metric Composer_."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "106780dbeffce346"
  },
  {
   "cell_type": "markdown",
   "source": [
    "To compose disparity metrics, the _Metric Composer_ should be applied. **Metric Composer** is responsible for the second stage of the model audit. Currently, it computes our custom error disparity, stability disparity, and uncertainty disparity metrics, but extending it for new disparity metrics is very simple. We noticed that more and more disparity metrics have appeared during the last decade, but most of them are based on the same group specific metrics. Hence, such a separation of group specific and disparity metrics computation allows us to experiment with different combinations of group specific metrics and avoid group metrics recomputation for a new set of disparity metrics."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "cee25ee4d75200ba"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Read the computed metrics from a file created by the metric computation interface."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "356161cf3fa140ce"
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "outputs": [],
   "source": [
    "models_metrics_dct = read_model_metric_dfs(SAVE_RESULTS_DIR_PATH, model_names=list(models_config.keys()))"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:24.921061Z",
     "start_time": "2024-06-02T14:03:24.887747Z"
    }
   },
   "id": "228c0e14df975b62"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Compose disparity metrics using _MetricsComposer_."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "8a597baac4d86943"
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "outputs": [],
   "source": [
    "metrics_composer = MetricsComposer(models_metrics_dct, config.sensitive_attributes_dct)\n",
    "models_composed_metrics_df = metrics_composer.compose_metrics()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:24.998457Z",
     "start_time": "2024-06-02T14:03:24.915871Z"
    }
   },
   "id": "fb9befb603132bb8"
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "outputs": [
    {
     "data": {
      "text/plain": "                              Metric       SEX     RAC1P  SEX&RAC1P  \\\n0                Accuracy_Difference  0.054206  0.009395   0.030921   \n1   Aleatoric_Uncertainty_Difference -0.052323 -0.043378  -0.062490   \n2        Aleatoric_Uncertainty_Ratio  0.932899  0.943534   0.918301   \n3   Epistemic_Uncertainty_Difference  0.001481 -0.008012  -0.005804   \n4        Epistemic_Uncertainty_Ratio  1.017297  0.909916   0.933545   \n..                               ...       ...       ...        ...   \n90                  Disparate_Impact  0.576996  0.530692   0.481141   \n91                    Std_Difference -0.008166 -0.004115  -0.009956   \n92                         Std_Ratio  0.829739  0.909165   0.782017   \n93                Equalized_Odds_TNR  0.080789  0.059151   0.061232   \n94                Equalized_Odds_TPR -0.085436 -0.196089  -0.187265   \n\n                Model_Name  \n0   DecisionTreeClassifier  \n1   DecisionTreeClassifier  \n2   DecisionTreeClassifier  \n3   DecisionTreeClassifier  \n4   DecisionTreeClassifier  \n..                     ...  \n90           XGBClassifier  \n91           XGBClassifier  \n92           XGBClassifier  \n93           XGBClassifier  \n94           XGBClassifier  \n\n[95 rows x 5 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>Metric</th>\n      <th>SEX</th>\n      <th>RAC1P</th>\n      <th>SEX&amp;RAC1P</th>\n      <th>Model_Name</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Accuracy_Difference</td>\n      <td>0.054206</td>\n      <td>0.009395</td>\n      <td>0.030921</td>\n      <td>DecisionTreeClassifier</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Aleatoric_Uncertainty_Difference</td>\n      <td>-0.052323</td>\n      <td>-0.043378</td>\n      <td>-0.062490</td>\n      <td>DecisionTreeClassifier</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Aleatoric_Uncertainty_Ratio</td>\n      <td>0.932899</td>\n      <td>0.943534</td>\n      <td>0.918301</td>\n      <td>DecisionTreeClassifier</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Epistemic_Uncertainty_Difference</td>\n      <td>0.001481</td>\n      <td>-0.008012</td>\n      <td>-0.005804</td>\n      <td>DecisionTreeClassifier</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Epistemic_Uncertainty_Ratio</td>\n      <td>1.017297</td>\n      <td>0.909916</td>\n      <td>0.933545</td>\n      <td>DecisionTreeClassifier</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>90</th>\n      <td>Disparate_Impact</td>\n      <td>0.576996</td>\n      <td>0.530692</td>\n      <td>0.481141</td>\n      <td>XGBClassifier</td>\n    </tr>\n    <tr>\n      <th>91</th>\n      <td>Std_Difference</td>\n      <td>-0.008166</td>\n      <td>-0.004115</td>\n      <td>-0.009956</td>\n      <td>XGBClassifier</td>\n    </tr>\n    <tr>\n      <th>92</th>\n      <td>Std_Ratio</td>\n      <td>0.829739</td>\n      <td>0.909165</td>\n      <td>0.782017</td>\n      <td>XGBClassifier</td>\n    </tr>\n    <tr>\n      <th>93</th>\n      <td>Equalized_Odds_TNR</td>\n      <td>0.080789</td>\n      <td>0.059151</td>\n      <td>0.061232</td>\n      <td>XGBClassifier</td>\n    </tr>\n    <tr>\n      <th>94</th>\n      <td>Equalized_Odds_TPR</td>\n      <td>-0.085436</td>\n      <td>-0.196089</td>\n      <td>-0.187265</td>\n      <td>XGBClassifier</td>\n    </tr>\n  </tbody>\n</table>\n<p>95 rows × 5 columns</p>\n</div>"
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "models_composed_metrics_df"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:25.000035Z",
     "start_time": "2024-06-02T14:03:24.955405Z"
    }
   },
   "id": "2025eaa778c0012b"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 6**: Create static visualizations using _Metric Static Visualizer_."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "48224b47306d8c85"
  },
  {
   "cell_type": "markdown",
   "source": [
    "**Metric Static Visualizer** allows us to build static visualizations for the computed metrics. It unifies different preprocessing methods for the computed  metrics and creates various data formats required for visualizations. Hence, users can simply call methods of the _MetricsVisualizer_ class and get custom plots for diverse metric analysis."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "64d8b9fcb30a61dd"
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "435b9d98",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:25.083646Z",
     "start_time": "2024-06-02T14:03:24.982838Z"
    }
   },
   "outputs": [],
   "source": [
    "visualizer = MetricsVisualizer(models_metrics_dct, models_composed_metrics_df, config.dataset_name,\n",
    "                               model_names=list(models_config.keys()),\n",
    "                               sensitive_attributes_dct=config.sensitive_attributes_dct)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "5efb1bf2",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:25.117998Z",
     "start_time": "2024-06-02T14:03:25.043021Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": "\n<div id=\"altair-viz-9bb3aa762f5c410fab22e082860b127e\"></div>\n<script type=\"text/javascript\">\n  var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n  (function(spec, embedOpt){\n    let outputDiv = document.currentScript.previousElementSibling;\n    if (outputDiv.id !== \"altair-viz-9bb3aa762f5c410fab22e082860b127e\") {\n      outputDiv = document.getElementById(\"altair-viz-9bb3aa762f5c410fab22e082860b127e\");\n    }\n    const paths = {\n      \"vega\": \"https://cdn.jsdelivr.net/npm//vega@5?noext\",\n      \"vega-lib\": \"https://cdn.jsdelivr.net/npm//vega-lib?noext\",\n      \"vega-lite\": \"https://cdn.jsdelivr.net/npm//vega-lite@4.17.0?noext\",\n      \"vega-embed\": \"https://cdn.jsdelivr.net/npm//vega-embed@6?noext\",\n    };\n\n    function maybeLoadScript(lib, version) {\n      var key = `${lib.replace(\"-\", \"\")}_version`;\n      return (VEGA_DEBUG[key] == version) ?\n        Promise.resolve(paths[lib]) :\n        new Promise(function(resolve, reject) {\n          var s = document.createElement('script');\n          document.getElementsByTagName(\"head\")[0].appendChild(s);\n          s.async = true;\n          s.onload = () => {\n            VEGA_DEBUG[key] = version;\n            return resolve(paths[lib]);\n          };\n          s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n          s.src = paths[lib];\n        });\n    }\n\n    function showError(err) {\n      outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n      throw err;\n    }\n\n    function displayChart(vegaEmbed) {\n      vegaEmbed(outputDiv, spec, embedOpt)\n        .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n    }\n\n    if(typeof define === \"function\" && define.amd) {\n      requirejs.config({paths});\n      require([\"vega-embed\"], displayChart, err => showError(`Error loading script: ${err.message}`));\n    } else {\n      maybeLoadScript(\"vega\", \"5\")\n        .then(() => maybeLoadScript(\"vega-lite\", \"4.17.0\"))\n        .then(() => maybeLoadScript(\"vega-embed\", \"6\"))\n        .catch(showError)\n        .then(() => displayChart(vegaEmbed));\n    }\n  })({\"config\": {\"view\": {\"continuousWidth\": 400, \"continuousHeight\": 300}, \"axis\": {\"labelFontSize\": 16, \"titleFontSize\": 20}, \"headerRow\": {\"labelAlign\": \"left\", \"labelAngle\": 0, \"labelFontSize\": 16, \"labelPadding\": 10, \"titleFontSize\": 20}}, \"data\": {\"name\": \"data-31b5d2aff8f13725df9faa8815629a65\"}, \"mark\": \"bar\", \"encoding\": {\"color\": {\"field\": \"model_name\", \"legend\": {\"labelFontSize\": 15, \"labelLimit\": 300, \"title\": \"Model Name\", \"titleFontSize\": 15, \"titleLimit\": 300}, \"scale\": {\"scheme\": \"tableau20\"}, \"type\": \"nominal\"}, \"row\": {\"field\": \"metric\", \"sort\": [\"Accuracy\", \"F1\", \"Label_Stability\", \"Std\", \"Aleatoric_Uncertainty\", \"Overall_Uncertainty\"], \"title\": \"Overall Metrics\", \"type\": \"nominal\"}, \"x\": {\"axis\": {\"grid\": true}, \"field\": \"overall\", \"title\": \"\", \"type\": \"quantitative\"}, \"y\": {\"axis\": null, \"field\": \"model_name\", \"type\": \"nominal\"}}, \"height\": 50, \"width\": 500, \"$schema\": \"https://vega.github.io/schema/vega-lite/v4.17.0.json\", \"datasets\": {\"data-31b5d2aff8f13725df9faa8815629a65\": [{\"overall\": 0.7833333333333333, \"metric\": \"Accuracy\", \"model_name\": \"DecisionTreeClassifier\"}, {\"overall\": 0.7541941175794297, \"metric\": \"Aleatoric_Uncertainty\", \"model_name\": \"DecisionTreeClassifier\"}, {\"overall\": 0.5997536945812808, \"metric\": \"F1\", \"model_name\": \"DecisionTreeClassifier\"}, {\"overall\": 0.6479666666666667, \"metric\": \"Label_Stability\", \"model_name\": \"DecisionTreeClassifier\"}, {\"overall\": 0.8405417435225854, \"metric\": \"Overall_Uncertainty\", \"model_name\": \"DecisionTreeClassifier\"}, {\"overall\": 0.1474064264508181, \"metric\": \"Std\", \"model_name\": \"DecisionTreeClassifier\"}, {\"overall\": 0.8206666666666667, \"metric\": \"Accuracy\", \"model_name\": \"LogisticRegression\"}, {\"overall\": 0.5674122826143212, \"metric\": \"Aleatoric_Uncertainty\", \"model_name\": \"LogisticRegression\"}, {\"overall\": 0.7304609218436874, \"metric\": \"F1\", \"model_name\": \"LogisticRegression\"}, {\"overall\": 0.9017333333333332, \"metric\": \"Label_Stability\", \"model_name\": \"LogisticRegression\"}, {\"overall\": 0.5838926669781963, \"metric\": \"Overall_Uncertainty\", \"model_name\": \"LogisticRegression\"}, {\"overall\": 0.0534208840332686, \"metric\": \"Std\", \"model_name\": \"LogisticRegression\"}, {\"overall\": 0.7723333333333333, \"metric\": \"Accuracy\", \"model_name\": \"KNeighborsClassifier\"}, {\"overall\": 0.4280561241139218, \"metric\": \"Aleatoric_Uncertainty\", \"model_name\": \"KNeighborsClassifier\"}, {\"overall\": 0.6647029945999018, \"metric\": \"F1\", \"model_name\": \"KNeighborsClassifier\"}, {\"overall\": 0.7470333333333334, \"metric\": \"Label_Stability\", \"model_name\": \"KNeighborsClassifier\"}, {\"overall\": 0.564651603641485, \"metric\": \"Overall_Uncertainty\", \"model_name\": \"KNeighborsClassifier\"}, {\"overall\": 0.1477692480854319, \"metric\": \"Std\", \"model_name\": \"KNeighborsClassifier\"}, {\"overall\": 0.7933333333333333, \"metric\": \"Accuracy\", \"model_name\": \"RandomForestClassifier\"}, {\"overall\": 0.6439822926849648, \"metric\": \"Aleatoric_Uncertainty\", \"model_name\": \"RandomForestClassifier\"}, {\"overall\": 0.6698615548455804, \"metric\": \"F1\", \"model_name\": \"RandomForestClassifier\"}, {\"overall\": 0.8994666666666667, \"metric\": \"Label_Stability\", \"model_name\": \"RandomForestClassifier\"}, {\"overall\": 0.6627584994515265, \"metric\": \"Overall_Uncertainty\", \"model_name\": \"RandomForestClassifier\"}, {\"overall\": 0.0594613644676053, \"metric\": \"Std\", \"model_name\": \"RandomForestClassifier\"}, {\"overall\": 0.8016666666666666, \"metric\": \"Accuracy\", \"model_name\": \"XGBClassifier\"}, {\"overall\": 0.660909116268158, \"metric\": \"Aleatoric_Uncertainty\", \"model_name\": \"XGBClassifier\"}, {\"overall\": 0.6876640419947506, \"metric\": \"F1\", \"model_name\": \"XGBClassifier\"}, {\"overall\": 0.9121, \"metric\": \"Label_Stability\", \"model_name\": \"XGBClassifier\"}, {\"overall\": 0.6707066243357207, \"metric\": \"Overall_Uncertainty\", \"model_name\": \"XGBClassifier\"}, {\"overall\": 0.0439709015190601, \"metric\": \"Std\", \"model_name\": \"XGBClassifier\"}]}}, {\"mode\": \"vega-lite\"});\n</script>",
      "text/plain": "alt.Chart(...)"
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visualizer.create_overall_metrics_bar_char(\n",
    "    metric_names=['Accuracy', 'F1', 'Label_Stability', 'Std', 'Aleatoric_Uncertainty', 'Overall_Uncertainty'],\n",
    "    plot_title=\"Overall Metrics\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 1200x900 with 2 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualizer.create_disparity_metric_heatmap(\n",
    "    model_names=list(models_params_for_tuning.keys()),\n",
    "    metrics_lst=[\n",
    "        # Error disparity metrics\n",
    "        'Equalized_Odds_TPR',\n",
    "        'Equalized_Odds_FPR',\n",
    "        'Disparate_Impact',\n",
    "        # Stability disparity metrics\n",
    "        'Label_Stability_Difference',\n",
    "        'Std_Ratio',\n",
    "        # Uncertainty disparity metrics\n",
    "        'Aleatoric_Uncertainty_Difference',\n",
    "    ],\n",
    "    groups_lst=config.sensitive_attributes_dct.keys(),\n",
    "    tolerance=0.005,\n",
    "    figsize_scale=(0.6, 0.5),\n",
    "    font_increase=-5,\n",
    ")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:25.594230Z",
     "start_time": "2024-06-02T14:03:25.084524Z"
    }
   },
   "id": "4bfca852f6ef428a"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## **Step 7**: Build an interactive web app using _Metric Interactive Visualizer_."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "58c76b42d4a482d2"
  },
  {
   "cell_type": "markdown",
   "source": [
    "Finally, user can apply _Metric Interactive Visualizer_ to create an interactive web application that guides responsible model selection process and generates a nutritional label for the selected model."
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "297f5a077817f19e"
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "2326c129",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-06-02T14:03:25.712910Z",
     "start_time": "2024-06-02T14:03:25.575076Z"
    }
   },
   "outputs": [],
   "source": [
    "interactive_metrics_visualizer = MetricsInteractiveVisualizer(X_data=data_loader.X_data,\n",
    "                                                              y_data=data_loader.y_data,\n",
    "                                                              model_metrics=models_metrics_dct,\n",
    "                                                              sensitive_attributes_dct=config.sensitive_attributes_dct)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "outputs": [],
   "source": [
    "interactive_metrics_visualizer.create_web_app()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:05:47.795721Z",
     "start_time": "2024-06-02T14:05:47.759941Z"
    }
   },
   "id": "7dcd9cca74e4c27c"
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-06-02T14:05:33.504432Z",
     "start_time": "2024-06-02T14:05:33.500399Z"
    }
   },
   "id": "5992ca0fc2dcdce7"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}