viz_testing_notebook.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-08-18T18:07:31.068996Z",
     "start_time": "2020-08-18T18:07:11.042342Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from aequitas.group import Group\n",
    "from aequitas.bias import Bias\n",
    "from aequitas import plot\n",
    "import pkg_resources"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pkg_resources.get_distribution(\"aequitas\").version"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-08-18T18:07:31.135890Z",
     "start_time": "2020-08-18T18:07:31.071760Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"./examples/data/compas_for_aequitas.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "g = Group()\n",
    "\n",
    "# Set the attr_cols explicitly to avoid a bug in Aequitas 0.41.0 (GitHub version):\n",
    "xtab, _ = g.get_crosstabs(df, attr_cols=[\"race\", \"sex\", \"age_cat\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "xtab.query(\"attribute_name == 'race'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "absolute_metrics = g.list_absolute_metrics(xtab)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2020-08-18T18:07:31.157816Z",
     "start_time": "2020-08-18T18:07:31.142203Z"
    }
   },
   "outputs": [],
   "source": [
    "b = Bias()\n",
    "majority_bdf = b.get_disparity_predefined_groups(\n",
    "    xtab,\n",
    "    original_df=df,\n",
    "    ref_groups_dict={\"race\": \"Caucasian\", \"sex\": \"Male\", \"age_cat\": \"25 - 45\"},\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# METRICS = [\"fpr\", \"fdr\", \"tpr\"]\n",
    "METRICS = [\"fpr\", \"fdr\"]\n",
    "# METRICS = [\"fpr\"]\n",
    "ATTRIBUTE = \"race\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.summary(\n",
    "    majority_bdf, attributes_list=[\"race\", \"sex\", \"age_cat\"], metrics_list=METRICS\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.disparity(majority_bdf, METRICS, \"age_cat\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.disparity(majority_bdf, METRICS, ATTRIBUTE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.disparity(majority_bdf, METRICS, ATTRIBUTE, accessibility_mode=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.absolute(majority_bdf, METRICS, ATTRIBUTE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.absolute(majority_bdf, METRICS, ATTRIBUTE, accessibility_mode=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.disparity_absolute(majority_bdf, METRICS, ATTRIBUTE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.disparity_absolute(majority_bdf, METRICS, ATTRIBUTE, accessibility_mode=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.scatter(majority_bdf, METRICS[0], METRICS[1], ATTRIBUTE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot.scatter(majority_bdf, METRICS[0], METRICS[1], ATTRIBUTE, accessibility_mode=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  }
 ],
 "metadata": {
  "hide_input": false,
  "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": "3.7.12"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}