Skip to content

Instantly share code, notes, and snippets.

@mahmoodm2

mahmoodm2/synthetic-data-generation-using-gan.ipynb Secret

Last active January 1, 2023 22:58
Show Gist options
  • Save mahmoodm2/6a18161ade47f7d53880332b43d516f0 to your computer and use it in GitHub Desktop.
Save mahmoodm2/6a18161ade47f7d53880332b43d516f0 to your computer and use it in GitHub Desktop.
Download ZIP
Synthetic data generation using GAN
Display the source blob
Display the rendered blob
Raw
synthetic-data-generation-using-gan.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "Synthetic data generation using GAN",
"provenance": [],
"collapsed_sections": [],
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"accelerator": "GPU"
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/mahmoodm2/6a18161ade47f7d53880332b43d516f0/synthetic-data-generation-using-gan.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6uKM9n6k5FIO"
},
"source": [
"###This Jupyter Notebook is used to demonstrare using the Generative Adversarial Netwrok(GAN) to generate synthetic tabular data."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "MkdDSNyY4-za"
},
"source": [
"###Downloading the Dataset"
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "Zz309b-ftcIO",
"outputId": "99f8a832-02fb-4243-f019-bfa946f3aae4"
},
"source": [
"!wget https://storage.googleapis.com/synthea-public/synthea_sample_data_csv_apr2020.zip\n",
"!unzip synthea_sample_data_csv_apr2020.zip"
],
"execution_count": 1,
"outputs": [
{
"output_type": "stream",
"text": [
"--2021-05-24 05:56:54-- https://storage.googleapis.com/synthea-public/synthea_sample_data_csv_apr2020.zip\n",
"Resolving storage.googleapis.com (storage.googleapis.com)... 142.250.128.128, 142.251.6.128, 74.125.126.128, ...\n",
"Connecting to storage.googleapis.com (storage.googleapis.com)|142.250.128.128|:443... connected.\n",
"HTTP request sent, awaiting response... 200 OK\n",
"Length: 8982431 (8.6M) [application/zip]\n",
"Saving to: ‘synthea_sample_data_csv_apr2020.zip.1’\n",
"\n",
"\r synthea_s 0%[ ] 0 --.-KB/s \rsynthea_sample_data 100%[===================>] 8.57M --.-KB/s in 0.05s \n",
"\n",
"2021-05-24 05:56:54 (183 MB/s) - ‘synthea_sample_data_csv_apr2020.zip.1’ saved [8982431/8982431]\n",
"\n",
"Archive: synthea_sample_data_csv_apr2020.zip\n",
"replace csv/medications.csv? [y]es, [n]o, [A]ll, [N]one, [r]ename: A\n",
" inflating: csv/medications.csv \n",
" inflating: csv/providers.csv \n",
" inflating: csv/payer_transitions.csv \n",
" inflating: csv/imaging_studies.csv \n",
" inflating: csv/supplies.csv \n",
" inflating: csv/payers.csv \n",
" inflating: csv/allergies.csv \n",
" inflating: csv/procedures.csv \n",
" inflating: csv/organizations.csv \n",
" inflating: csv/conditions.csv \n",
" inflating: csv/careplans.csv \n",
" inflating: csv/encounters.csv \n",
" inflating: csv/devices.csv \n",
" inflating: csv/immunizations.csv \n",
" inflating: csv/patients.csv \n",
" inflating: csv/observations.csv \n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "gi0jEinivij2",
"colab": {
"base_uri": "https://localhost:8080/"
},
"outputId": "2df905bf-17b9-4502-80bc-f9b4e30976ff"
},
"source": [
"!pip install ctgan"
],
"execution_count": 2,
"outputs": [
{
"output_type": "stream",
"text": [
"Requirement already satisfied: ctgan in /usr/local/lib/python3.7/dist-packages (0.4.2)\n",
"Requirement already satisfied: torch<2,>=1.4 in /usr/local/lib/python3.7/dist-packages (from ctgan) (1.8.1+cu101)\n",
"Requirement already satisfied: rdt<0.5,>=0.4.1 in /usr/local/lib/python3.7/dist-packages (from ctgan) (0.4.1)\n",
"Requirement already satisfied: numpy<2,>=1.18.0 in /usr/local/lib/python3.7/dist-packages (from ctgan) (1.19.5)\n",
"Requirement already satisfied: packaging in /usr/local/lib/python3.7/dist-packages (from ctgan) (20.9)\n",
"Requirement already satisfied: torchvision<1,>=0.5.0 in /usr/local/lib/python3.7/dist-packages (from ctgan) (0.9.1+cu101)\n",
"Requirement already satisfied: pandas<1.1.5,>=1.1 in /usr/local/lib/python3.7/dist-packages (from ctgan) (1.1.4)\n",
"Requirement already satisfied: scikit-learn<1,>=0.23 in /usr/local/lib/python3.7/dist-packages (from ctgan) (0.24.2)\n",
"Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torch<2,>=1.4->ctgan) (3.7.4.3)\n",
"Requirement already satisfied: scipy<2,>=1.4.1 in /usr/local/lib/python3.7/dist-packages (from rdt<0.5,>=0.4.1->ctgan) (1.4.1)\n",
"Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging->ctgan) (2.4.7)\n",
"Requirement already satisfied: pillow>=4.1.1 in /usr/local/lib/python3.7/dist-packages (from torchvision<1,>=0.5.0->ctgan) (7.1.2)\n",
"Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas<1.1.5,>=1.1->ctgan) (2.8.1)\n",
"Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas<1.1.5,>=1.1->ctgan) (2018.9)\n",
"Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn<1,>=0.23->ctgan) (1.0.1)\n",
"Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn<1,>=0.23->ctgan) (2.1.0)\n",
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas<1.1.5,>=1.1->ctgan) (1.15.0)\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "AVMSMTEKtojo",
"outputId": "ece02cd4-066f-4d79-8300-3bbc7037febd"
},
"source": [
"import pandas as pd\n",
"import warnings\n",
"warnings.filterwarnings('ignore')\n",
"\n",
"data = pd.read_csv('csv/patients.csv')\n",
"\n",
"keep_features = ['MARITAL', 'RACE', 'ETHNICITY', 'GENDER', 'BIRTHPLACE', 'CITY', 'ZIP', 'STATE' , 'HEALTHCARE_EXPENSES', 'HEALTHCARE_COVERAGE']\n",
"categorical_features = ['MARITAL', 'RACE', 'ETHNICITY', 'GENDER', 'BIRTHPLACE', 'CITY', 'STATE' , 'ZIP']\n",
"\n",
"real_data = data[keep_features]\n",
"print(real_data.columns)"
],
"execution_count": 3,
"outputs": [
{
"output_type": "stream",
"text": [
"Index(['MARITAL', 'RACE', 'ETHNICITY', 'GENDER', 'BIRTHPLACE', 'CITY', 'ZIP',\n",
" 'STATE', 'HEALTHCARE_EXPENSES', 'HEALTHCARE_COVERAGE'],\n",
" dtype='object')\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 415
},
"id": "1ExnblkT9Erh",
"outputId": "8bd18ca4-7aae-482b-d407-2b1f0b96f4a0"
},
"source": [
"real_data"
],
"execution_count": 4,
"outputs": [
{
"output_type": "execute_result",
"data": {
"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>MARITAL</th>\n",
" <th>RACE</th>\n",
" <th>ETHNICITY</th>\n",
" <th>GENDER</th>\n",
" <th>BIRTHPLACE</th>\n",
" <th>CITY</th>\n",
" <th>ZIP</th>\n",
" <th>STATE</th>\n",
" <th>HEALTHCARE_EXPENSES</th>\n",
" <th>HEALTHCARE_COVERAGE</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>hispanic</td>\n",
" <td>M</td>\n",
" <td>Marigot Saint Andrew Parish DM</td>\n",
" <td>Chicopee</td>\n",
" <td>1013.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>271227.08</td>\n",
" <td>1334.88</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>M</td>\n",
" <td>Danvers Massachusetts US</td>\n",
" <td>Somerville</td>\n",
" <td>2143.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>793946.01</td>\n",
" <td>3204.49</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>M</td>\n",
" <td>Springfield Massachusetts US</td>\n",
" <td>Chicopee</td>\n",
" <td>1020.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>574111.90</td>\n",
" <td>2606.40</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Yarmouth Massachusetts US</td>\n",
" <td>Lowell</td>\n",
" <td>1851.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>935630.30</td>\n",
" <td>8756.19</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>NaN</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>M</td>\n",
" <td>Patras Achaea GR</td>\n",
" <td>Boston</td>\n",
" <td>2135.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>598763.07</td>\n",
" <td>3772.20</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1166</th>\n",
" <td>M</td>\n",
" <td>asian</td>\n",
" <td>hispanic</td>\n",
" <td>F</td>\n",
" <td>Juarez Chihuahua MX</td>\n",
" <td>Cambridge</td>\n",
" <td>2141.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>1622314.87</td>\n",
" <td>32086.31</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1167</th>\n",
" <td>S</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>M</td>\n",
" <td>Upton Massachusetts US</td>\n",
" <td>Beverly</td>\n",
" <td>1915.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>979724.25</td>\n",
" <td>3130.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1168</th>\n",
" <td>S</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Fall River Massachusetts US</td>\n",
" <td>Norwood</td>\n",
" <td>NaN</td>\n",
" <td>Massachusetts</td>\n",
" <td>1560540.35</td>\n",
" <td>52391.24</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1169</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Springfield Massachusetts US</td>\n",
" <td>Norwood</td>\n",
" <td>2062.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>1375833.47</td>\n",
" <td>13157.00</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1170</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Worcester Massachusetts US</td>\n",
" <td>Norwood</td>\n",
" <td>2090.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>1510158.34</td>\n",
" <td>26565.65</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>1171 rows × 10 columns</p>\n",
"</div>"
],
"text/plain": [
" MARITAL RACE ... HEALTHCARE_EXPENSES HEALTHCARE_COVERAGE\n",
"0 M white ... 271227.08 1334.88\n",
"1 M white ... 793946.01 3204.49\n",
"2 M white ... 574111.90 2606.40\n",
"3 M white ... 935630.30 8756.19\n",
"4 NaN white ... 598763.07 3772.20\n",
"... ... ... ... ... ...\n",
"1166 M asian ... 1622314.87 32086.31\n",
"1167 S white ... 979724.25 3130.52\n",
"1168 S white ... 1560540.35 52391.24\n",
"1169 M white ... 1375833.47 13157.00\n",
"1170 M white ... 1510158.34 26565.65\n",
"\n",
"[1171 rows x 10 columns]"
]
},
"metadata": {
"tags": []
},
"execution_count": 4
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "RB4ukzuX3RXU"
},
"source": [
"## Training the model\n"
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "cpV1FWHevaWO",
"outputId": "c99b6c52-4072-4baf-e467-2ff933765304"
},
"source": [
"from ctgan import CTGANSynthesizer\n",
"\n",
"ctgan = CTGANSynthesizer(verbose=True)\n",
"ctgan.fit(real_data, categorical_features, epochs = 300)"
],
"execution_count": 5,
"outputs": [
{
"output_type": "stream",
"text": [
"Epoch 1, Loss G: 2.7220,Loss D: -0.0167\n",
"Epoch 2, Loss G: 2.6461,Loss D: -0.0450\n",
"Epoch 3, Loss G: 2.4273,Loss D: -0.0735\n",
"Epoch 4, Loss G: 2.7161,Loss D: -0.1428\n",
"Epoch 5, Loss G: 2.5028,Loss D: -0.1914\n",
"Epoch 6, Loss G: 2.5397,Loss D: -0.2294\n",
"Epoch 7, Loss G: 2.5450,Loss D: -0.2382\n",
"Epoch 8, Loss G: 2.5592,Loss D: -0.2775\n",
"Epoch 9, Loss G: 2.5199,Loss D: -0.3047\n",
"Epoch 10, Loss G: 2.5094,Loss D: -0.2992\n",
"Epoch 11, Loss G: 2.4861,Loss D: -0.3178\n",
"Epoch 12, Loss G: 2.4973,Loss D: -0.2785\n",
"Epoch 13, Loss G: 2.5476,Loss D: -0.3551\n",
"Epoch 14, Loss G: 2.3788,Loss D: -0.2778\n",
"Epoch 15, Loss G: 2.4980,Loss D: -0.2382\n",
"Epoch 16, Loss G: 2.2980,Loss D: -0.2273\n",
"Epoch 17, Loss G: 2.5093,Loss D: -0.3117\n",
"Epoch 18, Loss G: 2.3282,Loss D: -0.2865\n",
"Epoch 19, Loss G: 2.5446,Loss D: -0.3035\n",
"Epoch 20, Loss G: 2.3724,Loss D: -0.3000\n",
"Epoch 21, Loss G: 2.2893,Loss D: -0.3692\n",
"Epoch 22, Loss G: 2.4974,Loss D: -0.3598\n",
"Epoch 23, Loss G: 2.5267,Loss D: -0.3151\n",
"Epoch 24, Loss G: 2.4257,Loss D: -0.4370\n",
"Epoch 25, Loss G: 2.4397,Loss D: -0.3936\n",
"Epoch 26, Loss G: 2.5698,Loss D: -0.4311\n",
"Epoch 27, Loss G: 2.4059,Loss D: -0.4092\n",
"Epoch 28, Loss G: 2.6675,Loss D: -0.4984\n",
"Epoch 29, Loss G: 2.6464,Loss D: -0.4963\n",
"Epoch 30, Loss G: 2.3802,Loss D: -0.5488\n",
"Epoch 31, Loss G: 2.5288,Loss D: -0.4798\n",
"Epoch 32, Loss G: 2.4125,Loss D: -0.5222\n",
"Epoch 33, Loss G: 2.4611,Loss D: -0.4862\n",
"Epoch 34, Loss G: 2.3612,Loss D: -0.4904\n",
"Epoch 35, Loss G: 2.4119,Loss D: -0.4830\n",
"Epoch 36, Loss G: 2.2983,Loss D: -0.3992\n",
"Epoch 37, Loss G: 2.2561,Loss D: -0.3575\n",
"Epoch 38, Loss G: 2.1506,Loss D: -0.3187\n",
"Epoch 39, Loss G: 2.3781,Loss D: -0.4043\n",
"Epoch 40, Loss G: 2.0233,Loss D: -0.3109\n",
"Epoch 41, Loss G: 2.0642,Loss D: -0.3248\n",
"Epoch 42, Loss G: 2.0557,Loss D: -0.1483\n",
"Epoch 43, Loss G: 2.0589,Loss D: -0.1364\n",
"Epoch 44, Loss G: 1.9122,Loss D: -0.0944\n",
"Epoch 45, Loss G: 1.8047,Loss D: -0.2662\n",
"Epoch 46, Loss G: 1.7784,Loss D: -0.0055\n",
"Epoch 47, Loss G: 1.7492,Loss D: -0.0721\n",
"Epoch 48, Loss G: 1.5633,Loss D: -0.0962\n",
"Epoch 49, Loss G: 1.6252,Loss D: 0.0612\n",
"Epoch 50, Loss G: 1.3566,Loss D: 0.1531\n",
"Epoch 51, Loss G: 1.4798,Loss D: 0.1587\n",
"Epoch 52, Loss G: 1.3238,Loss D: 0.1388\n",
"Epoch 53, Loss G: 1.1748,Loss D: 0.2707\n",
"Epoch 54, Loss G: 1.2396,Loss D: 0.3116\n",
"Epoch 55, Loss G: 1.0063,Loss D: 0.4126\n",
"Epoch 56, Loss G: 1.2137,Loss D: 0.3624\n",
"Epoch 57, Loss G: 1.1719,Loss D: 0.3525\n",
"Epoch 58, Loss G: 1.1455,Loss D: 0.2201\n",
"Epoch 59, Loss G: 1.0129,Loss D: 0.3502\n",
"Epoch 60, Loss G: 1.2132,Loss D: 0.2884\n",
"Epoch 61, Loss G: 1.0484,Loss D: 0.3396\n",
"Epoch 62, Loss G: 1.2568,Loss D: 0.1721\n",
"Epoch 63, Loss G: 1.0787,Loss D: 0.3294\n",
"Epoch 64, Loss G: 1.2225,Loss D: 0.2763\n",
"Epoch 65, Loss G: 1.1407,Loss D: 0.2441\n",
"Epoch 66, Loss G: 1.2368,Loss D: 0.2397\n",
"Epoch 67, Loss G: 1.0690,Loss D: 0.2842\n",
"Epoch 68, Loss G: 1.3808,Loss D: 0.2167\n",
"Epoch 69, Loss G: 1.2878,Loss D: 0.1394\n",
"Epoch 70, Loss G: 1.3125,Loss D: 0.2123\n",
"Epoch 71, Loss G: 1.1866,Loss D: 0.1712\n",
"Epoch 72, Loss G: 1.1126,Loss D: 0.1258\n",
"Epoch 73, Loss G: 1.0783,Loss D: 0.1324\n",
"Epoch 74, Loss G: 1.2846,Loss D: 0.1744\n",
"Epoch 75, Loss G: 1.1055,Loss D: 0.1394\n",
"Epoch 76, Loss G: 1.1162,Loss D: 0.1411\n",
"Epoch 77, Loss G: 1.1037,Loss D: 0.1205\n",
"Epoch 78, Loss G: 1.1642,Loss D: 0.1726\n",
"Epoch 79, Loss G: 1.2163,Loss D: 0.1492\n",
"Epoch 80, Loss G: 1.3392,Loss D: 0.1172\n",
"Epoch 81, Loss G: 1.3620,Loss D: 0.0744\n",
"Epoch 82, Loss G: 1.2348,Loss D: 0.0469\n",
"Epoch 83, Loss G: 1.3188,Loss D: 0.0818\n",
"Epoch 84, Loss G: 1.2584,Loss D: 0.0977\n",
"Epoch 85, Loss G: 0.9544,Loss D: 0.0757\n",
"Epoch 86, Loss G: 1.3084,Loss D: 0.1224\n",
"Epoch 87, Loss G: 1.2146,Loss D: 0.0708\n",
"Epoch 88, Loss G: 1.3828,Loss D: 0.0582\n",
"Epoch 89, Loss G: 1.2468,Loss D: 0.0272\n",
"Epoch 90, Loss G: 0.9287,Loss D: 0.0418\n",
"Epoch 91, Loss G: 1.2695,Loss D: -0.0580\n",
"Epoch 92, Loss G: 1.1650,Loss D: 0.0311\n",
"Epoch 93, Loss G: 1.1005,Loss D: -0.0075\n",
"Epoch 94, Loss G: 1.0958,Loss D: -0.0779\n",
"Epoch 95, Loss G: 1.0175,Loss D: -0.0479\n",
"Epoch 96, Loss G: 1.0798,Loss D: -0.0387\n",
"Epoch 97, Loss G: 0.9573,Loss D: 0.0024\n",
"Epoch 98, Loss G: 1.2633,Loss D: 0.0258\n",
"Epoch 99, Loss G: 1.1980,Loss D: -0.0468\n",
"Epoch 100, Loss G: 0.8943,Loss D: -0.0297\n",
"Epoch 101, Loss G: 1.1320,Loss D: -0.0520\n",
"Epoch 102, Loss G: 1.1132,Loss D: 0.0590\n",
"Epoch 103, Loss G: 1.0539,Loss D: -0.0086\n",
"Epoch 104, Loss G: 0.9532,Loss D: -0.0071\n",
"Epoch 105, Loss G: 0.9822,Loss D: 0.0503\n",
"Epoch 106, Loss G: 0.9551,Loss D: 0.0449\n",
"Epoch 107, Loss G: 1.0087,Loss D: 0.0597\n",
"Epoch 108, Loss G: 0.9967,Loss D: -0.0227\n",
"Epoch 109, Loss G: 0.8712,Loss D: -0.0156\n",
"Epoch 110, Loss G: 1.0134,Loss D: 0.0081\n",
"Epoch 111, Loss G: 1.0755,Loss D: 0.0219\n",
"Epoch 112, Loss G: 0.7897,Loss D: -0.0476\n",
"Epoch 113, Loss G: 1.1207,Loss D: -0.0420\n",
"Epoch 114, Loss G: 1.0223,Loss D: 0.0482\n",
"Epoch 115, Loss G: 1.1407,Loss D: 0.0039\n",
"Epoch 116, Loss G: 1.0872,Loss D: -0.0750\n",
"Epoch 117, Loss G: 1.4322,Loss D: 0.0596\n",
"Epoch 118, Loss G: 1.2366,Loss D: -0.0371\n",
"Epoch 119, Loss G: 1.0306,Loss D: 0.0141\n",
"Epoch 120, Loss G: 0.8798,Loss D: 0.0237\n",
"Epoch 121, Loss G: 1.0133,Loss D: 0.0176\n",
"Epoch 122, Loss G: 1.0250,Loss D: 0.0203\n",
"Epoch 123, Loss G: 0.8172,Loss D: 0.1568\n",
"Epoch 124, Loss G: 0.9424,Loss D: 0.0468\n",
"Epoch 125, Loss G: 0.8411,Loss D: 0.1041\n",
"Epoch 126, Loss G: 0.9051,Loss D: 0.0677\n",
"Epoch 127, Loss G: 1.0352,Loss D: 0.0956\n",
"Epoch 128, Loss G: 1.1504,Loss D: 0.0208\n",
"Epoch 129, Loss G: 0.9771,Loss D: 0.0187\n",
"Epoch 130, Loss G: 1.1199,Loss D: 0.0049\n",
"Epoch 131, Loss G: 1.0768,Loss D: 0.0004\n",
"Epoch 132, Loss G: 0.9676,Loss D: -0.0217\n",
"Epoch 133, Loss G: 0.7691,Loss D: -0.0302\n",
"Epoch 134, Loss G: 0.8795,Loss D: -0.0011\n",
"Epoch 135, Loss G: 0.9124,Loss D: -0.0395\n",
"Epoch 136, Loss G: 1.0391,Loss D: -0.0147\n",
"Epoch 137, Loss G: 0.8113,Loss D: -0.0358\n",
"Epoch 138, Loss G: 0.8814,Loss D: -0.0458\n",
"Epoch 139, Loss G: 0.8053,Loss D: 0.0013\n",
"Epoch 140, Loss G: 0.8118,Loss D: 0.0303\n",
"Epoch 141, Loss G: 0.7793,Loss D: 0.0663\n",
"Epoch 142, Loss G: 0.8917,Loss D: 0.0327\n",
"Epoch 143, Loss G: 0.9252,Loss D: 0.0500\n",
"Epoch 144, Loss G: 0.8735,Loss D: 0.0354\n",
"Epoch 145, Loss G: 0.9865,Loss D: -0.0071\n",
"Epoch 146, Loss G: 1.0630,Loss D: 0.0339\n",
"Epoch 147, Loss G: 0.9280,Loss D: 0.0647\n",
"Epoch 148, Loss G: 1.0534,Loss D: -0.0397\n",
"Epoch 149, Loss G: 0.9826,Loss D: -0.0063\n",
"Epoch 150, Loss G: 0.9695,Loss D: 0.0367\n",
"Epoch 151, Loss G: 0.8811,Loss D: -0.0083\n",
"Epoch 152, Loss G: 0.8932,Loss D: 0.0303\n",
"Epoch 153, Loss G: 1.1060,Loss D: 0.0387\n",
"Epoch 154, Loss G: 0.9923,Loss D: 0.0255\n",
"Epoch 155, Loss G: 0.9793,Loss D: 0.1147\n",
"Epoch 156, Loss G: 0.7850,Loss D: 0.0505\n",
"Epoch 157, Loss G: 0.9503,Loss D: 0.0085\n",
"Epoch 158, Loss G: 0.9086,Loss D: 0.0088\n",
"Epoch 159, Loss G: 0.9582,Loss D: 0.0356\n",
"Epoch 160, Loss G: 1.0433,Loss D: -0.0400\n",
"Epoch 161, Loss G: 0.9563,Loss D: -0.0370\n",
"Epoch 162, Loss G: 0.8524,Loss D: -0.0236\n",
"Epoch 163, Loss G: 0.9272,Loss D: -0.0684\n",
"Epoch 164, Loss G: 0.9566,Loss D: -0.0119\n",
"Epoch 165, Loss G: 0.8715,Loss D: 0.0315\n",
"Epoch 166, Loss G: 0.7799,Loss D: 0.0613\n",
"Epoch 167, Loss G: 0.7510,Loss D: 0.0142\n",
"Epoch 168, Loss G: 0.9471,Loss D: -0.0258\n",
"Epoch 169, Loss G: 0.7951,Loss D: 0.0221\n",
"Epoch 170, Loss G: 0.7874,Loss D: -0.0423\n",
"Epoch 171, Loss G: 0.7759,Loss D: 0.0210\n",
"Epoch 172, Loss G: 0.6541,Loss D: 0.0298\n",
"Epoch 173, Loss G: 0.7881,Loss D: 0.0037\n",
"Epoch 174, Loss G: 0.7806,Loss D: 0.0232\n",
"Epoch 175, Loss G: 1.0056,Loss D: 0.0311\n",
"Epoch 176, Loss G: 0.7557,Loss D: 0.0046\n",
"Epoch 177, Loss G: 0.7037,Loss D: -0.0429\n",
"Epoch 178, Loss G: 0.9277,Loss D: -0.0689\n",
"Epoch 179, Loss G: 0.7939,Loss D: -0.0163\n",
"Epoch 180, Loss G: 0.8747,Loss D: 0.0225\n",
"Epoch 181, Loss G: 0.7599,Loss D: 0.0364\n",
"Epoch 182, Loss G: 1.0194,Loss D: 0.0112\n",
"Epoch 183, Loss G: 0.7092,Loss D: 0.0488\n",
"Epoch 184, Loss G: 0.7849,Loss D: 0.0450\n",
"Epoch 185, Loss G: 0.7199,Loss D: -0.0126\n",
"Epoch 186, Loss G: 0.7728,Loss D: 0.0037\n",
"Epoch 187, Loss G: 0.8458,Loss D: 0.0303\n",
"Epoch 188, Loss G: 0.8293,Loss D: -0.0230\n",
"Epoch 189, Loss G: 0.8573,Loss D: -0.0484\n",
"Epoch 190, Loss G: 0.8272,Loss D: 0.0031\n",
"Epoch 191, Loss G: 0.7993,Loss D: 0.0784\n",
"Epoch 192, Loss G: 0.7652,Loss D: -0.0231\n",
"Epoch 193, Loss G: 1.0990,Loss D: -0.0149\n",
"Epoch 194, Loss G: 0.6609,Loss D: 0.0494\n",
"Epoch 195, Loss G: 0.9004,Loss D: -0.0013\n",
"Epoch 196, Loss G: 0.9205,Loss D: -0.0169\n",
"Epoch 197, Loss G: 0.9881,Loss D: 0.0463\n",
"Epoch 198, Loss G: 0.7775,Loss D: 0.0623\n",
"Epoch 199, Loss G: 0.7577,Loss D: -0.0290\n",
"Epoch 200, Loss G: 0.4771,Loss D: 0.0394\n",
"Epoch 201, Loss G: 0.7217,Loss D: 0.0466\n",
"Epoch 202, Loss G: 0.7339,Loss D: 0.0388\n",
"Epoch 203, Loss G: 0.5890,Loss D: 0.0724\n",
"Epoch 204, Loss G: 0.6561,Loss D: 0.0555\n",
"Epoch 205, Loss G: 0.8715,Loss D: -0.0098\n",
"Epoch 206, Loss G: 0.7008,Loss D: 0.0330\n",
"Epoch 207, Loss G: 0.7343,Loss D: 0.0189\n",
"Epoch 208, Loss G: 0.7506,Loss D: -0.0591\n",
"Epoch 209, Loss G: 0.7358,Loss D: -0.1123\n",
"Epoch 210, Loss G: 0.8187,Loss D: -0.0802\n",
"Epoch 211, Loss G: 0.8247,Loss D: -0.0270\n",
"Epoch 212, Loss G: 0.7559,Loss D: -0.0096\n",
"Epoch 213, Loss G: 0.5888,Loss D: 0.1001\n",
"Epoch 214, Loss G: 0.5673,Loss D: 0.1054\n",
"Epoch 215, Loss G: 0.5811,Loss D: 0.0744\n",
"Epoch 216, Loss G: 0.9078,Loss D: 0.1532\n",
"Epoch 217, Loss G: 0.5786,Loss D: 0.1000\n",
"Epoch 218, Loss G: 0.7233,Loss D: 0.0323\n",
"Epoch 219, Loss G: 0.7319,Loss D: 0.0424\n",
"Epoch 220, Loss G: 0.7076,Loss D: 0.0082\n",
"Epoch 221, Loss G: 0.6885,Loss D: 0.0088\n",
"Epoch 222, Loss G: 0.7126,Loss D: -0.0506\n",
"Epoch 223, Loss G: 0.5396,Loss D: 0.0225\n",
"Epoch 224, Loss G: 0.5722,Loss D: -0.0346\n",
"Epoch 225, Loss G: 0.9440,Loss D: -0.0512\n",
"Epoch 226, Loss G: 0.8581,Loss D: -0.0096\n",
"Epoch 227, Loss G: 0.7515,Loss D: -0.0632\n",
"Epoch 228, Loss G: 0.5515,Loss D: -0.0164\n",
"Epoch 229, Loss G: 0.6719,Loss D: -0.0063\n",
"Epoch 230, Loss G: 0.4614,Loss D: 0.0704\n",
"Epoch 231, Loss G: 0.4597,Loss D: 0.0108\n",
"Epoch 232, Loss G: 0.5004,Loss D: 0.1508\n",
"Epoch 233, Loss G: 0.5043,Loss D: 0.0843\n",
"Epoch 234, Loss G: 0.6481,Loss D: -0.0396\n",
"Epoch 235, Loss G: 0.5557,Loss D: -0.0045\n",
"Epoch 236, Loss G: 0.7464,Loss D: -0.0184\n",
"Epoch 237, Loss G: 0.5849,Loss D: 0.0124\n",
"Epoch 238, Loss G: 0.5131,Loss D: -0.0530\n",
"Epoch 239, Loss G: 0.6377,Loss D: -0.0355\n",
"Epoch 240, Loss G: 0.5646,Loss D: 0.0026\n",
"Epoch 241, Loss G: 0.6161,Loss D: -0.1150\n",
"Epoch 242, Loss G: 0.4843,Loss D: -0.0011\n",
"Epoch 243, Loss G: 0.8724,Loss D: -0.0231\n",
"Epoch 244, Loss G: 0.5319,Loss D: -0.0043\n",
"Epoch 245, Loss G: 0.4672,Loss D: 0.0975\n",
"Epoch 246, Loss G: 0.5956,Loss D: 0.1880\n",
"Epoch 247, Loss G: 0.7126,Loss D: 0.0936\n",
"Epoch 248, Loss G: 0.4375,Loss D: 0.1400\n",
"Epoch 249, Loss G: 0.7449,Loss D: 0.0571\n",
"Epoch 250, Loss G: 0.4647,Loss D: 0.0067\n",
"Epoch 251, Loss G: 0.5160,Loss D: 0.0214\n",
"Epoch 252, Loss G: 0.7419,Loss D: -0.0534\n",
"Epoch 253, Loss G: 0.5988,Loss D: -0.0425\n",
"Epoch 254, Loss G: 0.3704,Loss D: -0.0227\n",
"Epoch 255, Loss G: 0.5754,Loss D: -0.0804\n",
"Epoch 256, Loss G: 0.5358,Loss D: 0.0334\n",
"Epoch 257, Loss G: 0.3127,Loss D: 0.0243\n",
"Epoch 258, Loss G: 0.2116,Loss D: 0.0720\n",
"Epoch 259, Loss G: 0.3320,Loss D: 0.0687\n",
"Epoch 260, Loss G: 0.2940,Loss D: 0.0422\n",
"Epoch 261, Loss G: 0.6401,Loss D: 0.0058\n",
"Epoch 262, Loss G: 0.4052,Loss D: 0.0804\n",
"Epoch 263, Loss G: 0.5086,Loss D: -0.0041\n",
"Epoch 264, Loss G: 0.4924,Loss D: 0.0162\n",
"Epoch 265, Loss G: 0.5532,Loss D: -0.0401\n",
"Epoch 266, Loss G: 0.5285,Loss D: -0.0381\n",
"Epoch 267, Loss G: 0.4629,Loss D: -0.0970\n",
"Epoch 268, Loss G: 0.6104,Loss D: -0.0922\n",
"Epoch 269, Loss G: 0.3659,Loss D: 0.0015\n",
"Epoch 270, Loss G: 0.4166,Loss D: 0.0393\n",
"Epoch 271, Loss G: 0.4172,Loss D: 0.0997\n",
"Epoch 272, Loss G: 0.2331,Loss D: 0.0999\n",
"Epoch 273, Loss G: 0.3405,Loss D: 0.0730\n",
"Epoch 274, Loss G: 0.3338,Loss D: 0.0279\n",
"Epoch 275, Loss G: 0.3571,Loss D: 0.0667\n",
"Epoch 276, Loss G: 0.3951,Loss D: 0.0140\n",
"Epoch 277, Loss G: 0.2518,Loss D: 0.0726\n",
"Epoch 278, Loss G: 0.3112,Loss D: -0.0216\n",
"Epoch 279, Loss G: 0.4161,Loss D: -0.0146\n",
"Epoch 280, Loss G: 0.3186,Loss D: -0.0345\n",
"Epoch 281, Loss G: 0.2794,Loss D: -0.0716\n",
"Epoch 282, Loss G: 0.3211,Loss D: -0.0570\n",
"Epoch 283, Loss G: 0.2710,Loss D: -0.0596\n",
"Epoch 284, Loss G: 0.3674,Loss D: -0.0093\n",
"Epoch 285, Loss G: 0.2425,Loss D: -0.0092\n",
"Epoch 286, Loss G: 0.3579,Loss D: 0.0631\n",
"Epoch 287, Loss G: 0.4247,Loss D: 0.0355\n",
"Epoch 288, Loss G: 0.2433,Loss D: -0.0335\n",
"Epoch 289, Loss G: 0.3719,Loss D: -0.0625\n",
"Epoch 290, Loss G: 0.4448,Loss D: 0.0788\n",
"Epoch 291, Loss G: 0.2918,Loss D: 0.0329\n",
"Epoch 292, Loss G: 0.3490,Loss D: 0.1048\n",
"Epoch 293, Loss G: 0.3431,Loss D: 0.0557\n",
"Epoch 294, Loss G: 0.2143,Loss D: -0.0053\n",
"Epoch 295, Loss G: 0.2369,Loss D: 0.1335\n",
"Epoch 296, Loss G: 0.1772,Loss D: 0.0921\n",
"Epoch 297, Loss G: 0.3189,Loss D: 0.0471\n",
"Epoch 298, Loss G: 0.1971,Loss D: 0.0551\n",
"Epoch 299, Loss G: 0.0577,Loss D: 0.0345\n",
"Epoch 300, Loss G: 0.2606,Loss D: 0.0089\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Z9TvE4Ae46vy"
},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Ij1MvT4Q3rDJ"
},
"source": [
"## Evaluation\n",
"\n",
"Uisng the sample() function of the trained GAN to generate fakes samples."
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 202
},
"id": "dAOZhYjp96NN",
"outputId": "7f04425a-51aa-40d2-83d2-20dc5f74611c"
},
"source": [
"fake_data = ctgan.sample(1000)\n",
"\n",
"fake_data.head()\n"
],
"execution_count": 6,
"outputs": [
{
"output_type": "execute_result",
"data": {
"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>MARITAL</th>\n",
" <th>RACE</th>\n",
" <th>ETHNICITY</th>\n",
" <th>GENDER</th>\n",
" <th>BIRTHPLACE</th>\n",
" <th>CITY</th>\n",
" <th>ZIP</th>\n",
" <th>STATE</th>\n",
" <th>HEALTHCARE_EXPENSES</th>\n",
" <th>HEALTHCARE_COVERAGE</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>NaN</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Caguas Puerto Rico PR</td>\n",
" <td>Chatham</td>\n",
" <td>2148.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>1.401285e+06</td>\n",
" <td>9040.259524</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>NaN</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Quincy Massachusetts US</td>\n",
" <td>Groveland</td>\n",
" <td>1106.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>6.068832e+05</td>\n",
" <td>10299.582375</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>S</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Holyoke Massachusetts US</td>\n",
" <td>Methuen</td>\n",
" <td>2114.0</td>\n",
" <td>Massachusetts</td>\n",
" <td>1.200787e+06</td>\n",
" <td>6400.250611</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>M</td>\n",
" <td>white</td>\n",
" <td>nonhispanic</td>\n",
" <td>M</td>\n",
" <td>Worcester Massachusetts US</td>\n",
" <td>Lynn</td>\n",
" <td>NaN</td>\n",
" <td>Massachusetts</td>\n",
" <td>1.294887e+04</td>\n",
" <td>11408.045525</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>S</td>\n",
" <td>black</td>\n",
" <td>nonhispanic</td>\n",
" <td>F</td>\n",
" <td>Dartmouth Massachusetts US</td>\n",
" <td>Lowell</td>\n",
" <td>NaN</td>\n",
" <td>Massachusetts</td>\n",
" <td>6.576685e+05</td>\n",
" <td>11668.114383</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" MARITAL RACE ... HEALTHCARE_EXPENSES HEALTHCARE_COVERAGE\n",
"0 NaN white ... 1.401285e+06 9040.259524\n",
"1 NaN white ... 6.068832e+05 10299.582375\n",
"2 S white ... 1.200787e+06 6400.250611\n",
"3 M white ... 1.294887e+04 11408.045525\n",
"4 S black ... 6.576685e+05 11668.114383\n",
"\n",
"[5 rows x 10 columns]"
]
},
"metadata": {
"tags": []
},
"execution_count": 6
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "9e1ilRTGz8_9"
},
"source": [
"\n",
"import seaborn as sns\n",
"import matplotlib.pyplot as plt"
],
"execution_count": 7,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"id": "LdG7lIppFETC"
},
"source": [
"### Comparing the Real and Fake datasets"
]
},
{
"cell_type": "code",
"metadata": {
"id": "94Cp1ZZ36-_I",
"outputId": "058bd62c-a509-4dd2-eac9-dd1a5675faa5",
"colab": {
"base_uri": "https://localhost:8080/"
}
},
"source": [
"!pip install dython"
],
"execution_count": 8,
"outputs": [
{
"output_type": "stream",
"text": [
"Requirement already satisfied: dython in /usr/local/lib/python3.7/dist-packages (0.6.5.post1)\n",
"Requirement already satisfied: seaborn in /usr/local/lib/python3.7/dist-packages (from dython) (0.11.1)\n",
"Requirement already satisfied: matplotlib in /usr/local/lib/python3.7/dist-packages (from dython) (3.2.2)\n",
"Requirement already satisfied: scikit-plot>=0.3.7 in /usr/local/lib/python3.7/dist-packages (from dython) (0.3.7)\n",
"Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from dython) (1.19.5)\n",
"Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from dython) (1.4.1)\n",
"Requirement already satisfied: scikit-learn in /usr/local/lib/python3.7/dist-packages (from dython) (0.24.2)\n",
"Requirement already satisfied: pandas>=0.23.4 in /usr/local/lib/python3.7/dist-packages (from dython) (1.1.4)\n",
"Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->dython) (2.4.7)\n",
"Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->dython) (1.3.1)\n",
"Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->dython) (2.8.1)\n",
"Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->dython) (0.10.0)\n",
"Requirement already satisfied: joblib>=0.10 in /usr/local/lib/python3.7/dist-packages (from scikit-plot>=0.3.7->dython) (1.0.1)\n",
"Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->dython) (2.1.0)\n",
"Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.23.4->dython) (2018.9)\n",
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.1->matplotlib->dython) (1.15.0)\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 473
},
"id": "dml4lOPKE_gS",
"outputId": "441cb864-e7af-409b-8530-d1a0a3e16cff"
},
"source": [
"from dython.nominal import compute_associations, numerical_encoding\n",
"\n",
"real_corr = compute_associations(real_data, nominal_columns=categorical_features, theil_u=True)\n",
"fake_corr = compute_associations(fake_data, nominal_columns=categorical_features, theil_u=True)\n",
"\n",
"fig, axes = plt.subplots(1, 2, figsize=(35, 8))\n",
"\n",
"sns.set(style=\"white\")\n",
"sns.heatmap(real_corr, ax=axes[0], vmax=.3, square=True, center=0,\n",
" linewidths=.5, cbar_kws={\"shrink\": .5}, fmt='.2f')\n",
"axes[0].set_title('Real', size=20)\n",
"\n",
"sns.heatmap(fake_corr, ax=axes[1], vmax=.3, square=True, center=0,\n",
" linewidths=.5, cbar_kws={\"shrink\": .5}, fmt='.2f')\n",
"axes[1].set_title('Fake', size=20)\n",
"\n",
"\n",
"for ax in axes:\n",
" ax.set_xticklabels(\n",
" ax.get_xticklabels(),\n",
" rotation=45,\n",
" horizontalalignment='right'\n",
" )\n",
"\n"
],
"execution_count": 9,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 2520x576 with 4 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 394
},
"id": "tf1K-qRSH4bp",
"outputId": "fa3deb46-3036-4666-bf02-0139bb38b4e9"
},
"source": [
"cont_data = [ c for c in real_data.columns if c not in categorical_features]\n",
"cols = 2\n",
"\n",
"rows =max (1 , len(cont_data) // cols )\n",
"\n",
"fig, ax= plt.subplots(rows, cols, figsize=(25, 6 * rows))\n",
"\n",
"axes = ax.flatten()\n",
"i=0\n",
"for col in cont_data: \n",
" sns.distplot(real_data[col], ax=axes[i], label='Real', color = 'blue') \n",
"\n",
" sns.distplot(fake_data[col], ax=axes[i], label='Fake', color = 'orange')\n",
" \n",
" axes[i].set_title(col, size=20)\n",
" axes[i].legend(loc=0,prop={'size': 20})\n",
"\n",
" i+=1"
],
"execution_count": 10,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1800x432 with 2 Axes>"
]
},
"metadata": {
"tags": []
}
}
]
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment