nb_issues_with_aml_js_distance_implementation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import random\n",
    "from scipy.spatial import distance\n",
    "from utils import jensen_shannon_distance_numerical\n",
    "from evidently.calculations.stattests import jensenshannon_stat_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def jensen_shannon_distance(production_col: pd.DataFrame, baseline_col: pd.DataFrame) -> float:\n",
    "    \"\"\"Compute data drift using Jensen-Shannon Distance for each feature.\n",
    "\n",
    "    Args:\n",
    "        production_col: DataFrame column from production dataset provided by user.\n",
    "        baseline_col: DataFrame column from baseline dataset provided by user.\n",
    "\n",
    "    Returns:\n",
    "        float: The Jensen-Shannon distances between production and baseline column\n",
    "\n",
    "    \"\"\"\n",
    "    values_per_column = list(set(baseline_col.unique()) | set(production_col.unique()))\n",
    "    current_frequencies = np.array([production_col.to_list().count(value) for value in values_per_column])\n",
    "    reference_frequencies = np.array([baseline_col.to_list().count(value) for value in values_per_column])\n",
    "    current_ratios = current_frequencies / np.sum(current_frequencies)\n",
    "    reference_ratios = reference_frequencies / np.sum(reference_frequencies)\n",
    "    return distance.jensenshannon(current_ratios, reference_ratios)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "observations = 10_000\n",
    "x = pd.Series(np.random.normal(50, 15, observations)) # reference\n",
    "w = pd.Series(np.random.normal(50, 15, observations)) # identical with x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Jensen-Shannon Distance Current Implementation: 0.8325546111576977\n"
     ]
    }
   ],
   "source": [
    "print(\"Jensen-Shannon Distance Current Implementation:\", jensen_shannon_distance(x, w))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Jensen-Shannon Distance Implementation in this Repo 0.016184214339995565\n"
     ]
    }
   ],
   "source": [
    "print(\"Jensen-Shannon Distance Implementation in this Repo\", jensen_shannon_distance_numerical(x, w, bin_strategy='stu'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Jensen Shannon Evidently Implementation:  StatTestResult(drift_score=0.018675562151804537, drifted=False, actual_threshold=0.1)\n"
     ]
    }
   ],
   "source": [
    "print(\"Jensen Shannon Evidently Implementation: \", jensenshannon_stat_test(x, w, feature_type='num', threshold=0.1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x.hist(bins=40, alpha=0.5)\n",
    "w.hist(bins=40, alpha=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "data-drift-env",
   "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": "3.8.15"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "de32243efe8733ca4a15d9aaf292021ac27ae74c6260a7ba037a76e64ea1f91c"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}