ml/ab_tests.ipynb

{
 "metadata": {
  "orig_nbformat": 2,
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3",
   "language": "python"
  },
  "interpreter": {
   "hash": "7fe315b7a207e63d5c2ef5139e719409788deb4436627680109abc8a609c6411"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2,
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "import json # will be needed for saving preprocessing details\r\n",
    "import numpy as np # for data manipulation\r\n",
    "import pandas as pd # for data manipulation\r\n",
    "from sklearn.model_selection import train_test_split # will be used for data split\r\n",
    "import requests\r\n",
    "\r\n",
    "df = pd.read_csv('zoo/data/german_data.csv', index_col=0)\r\n",
    "df = df.drop(columns = ['Saving accounts', 'Checking account'])\r\n",
    "\r\n",
    "x_cols = [c for c in df.columns if c != 'Risk']\r\n",
    "\r\n",
    "# set input matrix and target column\r\n",
    "X = df[x_cols]\r\n",
    "y = df['Risk']\r\n",
    "# show first rows of data\r\n",
    "\r\n",
    "df.head()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "   Age     Sex  Job Housing  Credit amount  Duration              Purpose  \\\n",
       "0   67    male    2     own           1169         6             radio/TV   \n",
       "1   22  female    2     own           5951        48             radio/TV   \n",
       "2   49    male    1     own           2096        12            education   \n",
       "3   45    male    2    free           7882        42  furniture/equipment   \n",
       "4   53    male    2    free           4870        24                  car   \n",
       "\n",
       "   Risk  \n",
       "0  good  \n",
       "1   bad  \n",
       "2  good  \n",
       "3  good  \n",
       "4   bad  "
      ],
      "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>Age</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Job</th>\n",
       "      <th>Housing</th>\n",
       "      <th>Credit amount</th>\n",
       "      <th>Duration</th>\n",
       "      <th>Purpose</th>\n",
       "      <th>Risk</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>67</td>\n",
       "      <td>male</td>\n",
       "      <td>2</td>\n",
       "      <td>own</td>\n",
       "      <td>1169</td>\n",
       "      <td>6</td>\n",
       "      <td>radio/TV</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22</td>\n",
       "      <td>female</td>\n",
       "      <td>2</td>\n",
       "      <td>own</td>\n",
       "      <td>5951</td>\n",
       "      <td>48</td>\n",
       "      <td>radio/TV</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>49</td>\n",
       "      <td>male</td>\n",
       "      <td>1</td>\n",
       "      <td>own</td>\n",
       "      <td>2096</td>\n",
       "      <td>12</td>\n",
       "      <td>education</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>45</td>\n",
       "      <td>male</td>\n",
       "      <td>2</td>\n",
       "      <td>free</td>\n",
       "      <td>7882</td>\n",
       "      <td>42</td>\n",
       "      <td>furniture/equipment</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>53</td>\n",
       "      <td>male</td>\n",
       "      <td>2</td>\n",
       "      <td>free</td>\n",
       "      <td>4870</td>\n",
       "      <td>24</td>\n",
       "      <td>car</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "metadata": {},
     "execution_count": 1
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "source": [
    "# data split train / test\r\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, random_state=1234)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "source": [
    "for i in range(2):\r\n",
    "    input_data = dict(X_test.iloc[i])\r\n",
    "    target = y_test.iloc[i]\r\n",
    "    r = requests.post(\"http://127.0.0.1:8000/api/v1/algorithms/predict?status=ab_testing\", input_data)\r\n",
    "    response = r.json()\r\n",
    "    print(response)\r\n",
    "    # provide feedback\r\n",
    "    requests.put(\"http://127.0.0.1:8000/api/v1/requests/{}\".format(response[\"request_id\"]), {\"feedback\": target})"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "{'probability': 0.05733167685353127, 'label': 'bad', 'status': 'OK', 'request_id': 1}\n",
      "{'probability': 0.43333333333333335, 'label': 'bad', 'status': 'OK', 'request_id': 2}\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "r = requests.post(\"http://127.0.0.1:8000/api/v1/abtest/predict?status=ab_testing\", input_data)"
   ],
   "outputs": [],
   "metadata": {}
  }
 ]
}