2nd-TU39882 (1) (1).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b6e49e42-6c3e-4eb7-a30d-b21a40868f8c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import division\n",
    "from __future__ import print_function\n",
    "from __future__ import absolute_import\n",
    "\n",
    "# Suppresses warnings\n",
    "# Warnings were noted prior to cleaning up the notebook, suppressing them are for aesthetics\n",
    "import warnings\n",
    "warnings.filterwarnings(action='ignore')\n",
    "\n",
    "import time\n",
    "import copy\n",
    "import random\n",
    "import tensorflow as tf\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import scipy.stats as stats\n",
    "from timeit import default_timer as timer\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import SGDClassifier, LogisticRegression\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from imblearn.metrics import classification_report_imbalanced\n",
    "from imblearn import pipeline\n",
    "from imblearn.combine import SMOTETomek\n",
    "from imblearn.over_sampling import SMOTE, RandomOverSampler\n",
    "from imblearn.under_sampling import TomekLinks\n",
    "tf.compat.v1.disable_eager_execution()\n",
    "# Initialise seed value - 42\n",
    "seed = 42\n",
    "random.seed(42)\n",
    "tf.random.set_seed(seed)\n",
    "np.random.seed(seed)\n",
    "sns.set_style(\"whitegrid\")\n",
    "\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8e669b8b-a4d1-4a87-8f80-e130f3ca7631",
   "metadata": {},
   "outputs": [],
   "source": [
    "#loading credit card fraud dataset to pandas datafame\n",
    "credit=pd.read_csv('creditcard.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6451fd0c-4d03-47d7-a776-6a7fc3d006c2",
   "metadata": {},
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9  ...       V21       V22       V23       V24       V25  \\\n",
       "0  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "1  0.085102 -0.255425  ... -0.225775 -0.638672  0.101288 -0.339846  0.167170   \n",
       "2  0.247676 -1.514654  ...  0.247998  0.771679  0.909412 -0.689281 -0.327642   \n",
       "3  0.377436 -1.387024  ... -0.108300  0.005274 -0.190321 -1.175575  0.647376   \n",
       "4 -0.270533  0.817739  ... -0.009431  0.798278 -0.137458  0.141267 -0.206010   \n",
       "\n",
       "        V26       V27       V28  Amount  Class  \n",
       "0 -0.189115  0.133558 -0.021053  149.62      0  \n",
       "1  0.125895 -0.008983  0.014724    2.69      0  \n",
       "2 -0.139097 -0.055353 -0.059752  378.66      0  \n",
       "3 -0.221929  0.062723  0.061458  123.50      0  \n",
       "4  0.502292  0.219422  0.215153   69.99      0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#show the first 5 rows of the dataset\n",
    "credit.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "5a0243db-0514-4442-849c-d521abdbca4b",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Generative Model Classes\n",
    "class GAN(object):\n",
    "    \n",
    "    \"\"\" Invoke the methods in sequential order according to the scope, otherwise there would be errors. \"\"\"\n",
    "    \n",
    "    def __init__(self, batches=2048, X_nodes=30, y_nodes=1, z_dims=100, X_name='X_data', z_name='z_prior', orig_loss_fn=True):\n",
    "        \n",
    "        # Initialise inputs\n",
    "        self.batch_, self.X_node_, self.y_node_, self.z_dim_ = batches, X_nodes, y_nodes, z_dims\n",
    "        self.original_loss_fn_ = orig_loss_fn\n",
    "        \n",
    "        # Initialise placeholders\n",
    "        self.X_ = tf.compat.v1.placeholder(tf.float32, [None, self.X_node_], name= X_name)\n",
    "        self.prior_z_ = tf.compat.v1.placeholder(tf.float32, [None, self.z_dim_], name= z_name)\n",
    "        \n",
    "        # Generator parameters\n",
    "        self.gen_X_, self.gen_params_ = None, None\n",
    "        \n",
    "        # Discriminator parameters\n",
    "        self.y_data_, self.gen_y_, self.dis_params_ = None, None, None\n",
    "        \n",
    "        # Optimisation parameters\n",
    "        self.d_loss_, self.g_loss_, self.d_trainer_, self.g_trainer_ = None, None, None, None\n",
    "        self.g2_, self.d2_, self.gen_d2_ = None, None, None\n",
    "                  \n",
    "    def generator(self, h1_nodes=500, h2_nodes=500):\n",
    "        \n",
    "        # Generator weights/biases\n",
    "        g_h1 = {'weights': tf.Variable(xavier_init([self.z_dim_, h1_nodes], 'g_w1', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros(h1_nodes), name='g_b1', dtype=tf.float32)}\n",
    "        g_h2 = {'weights': tf.Variable(xavier_init([h1_nodes, h2_nodes], 'g_w2', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([h2_nodes]), name='g_b2', dtype=tf.float32)}\n",
    "        g_h3 = {'weights': tf.Variable(xavier_init([h2_nodes, self.X_node_], 'g_w3', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([self.X_node_]), name='g_b3', dtype=tf.float32)}\n",
    "\n",
    "        # Fully connect the layers\n",
    "        g1 = tf.nn.leaky_relu(tf.matmul(self.prior_z_, g_h1['weights']) + g_h1['biases'])\n",
    "        if self.original_loss_fn_:\n",
    "            self.g2_ = tf.nn.leaky_relu(tf.matmul(g1, g_h2['weights']) + g_h2['biases'])\n",
    "        else:\n",
    "            self.g2_ = tf.matmul(g1, g_h2['weights']) + g_h2['biases']\n",
    "#         g_out = tf.nn.sigmoid(tf.matmul(self.g2_, g_h3['weights']) + g_h3['biases'])\n",
    "        g_out = tf.nn.tanh(tf.matmul(self.g2_, g_h3['weights']) + g_h3['biases'])\n",
    "\n",
    "        # Generate fake x's with from output layer of generator\n",
    "        self.gen_X_ = g_out\n",
    "\n",
    "        # List of 'var_list' for generator trainer to optimise\n",
    "        self.gen_params_ = [g_h1['weights'], g_h1['biases'],\n",
    "                           g_h2['weights'], g_h2['biases'],\n",
    "                           g_h3['weights'], g_h3['biases']]\n",
    "        \n",
    "    def discriminator(self, h1_nodes=500, h2_nodes=500):\n",
    "        \n",
    "        ## Discriminator weights/biases ##\n",
    "        d_h1 = {'weights': tf.Variable(xavier_init([self.X_node_, h1_nodes], 'd_w1', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([h1_nodes]), name='d_b1', dtype=tf.float32)}\n",
    "        d_h2 = {'weights': tf.Variable(xavier_init([h1_nodes, h2_nodes], 'd_w2', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([h2_nodes]), name='d_b2', dtype=tf.float32)}\n",
    "        d_h3 = {'weights': tf.Variable(xavier_init([h2_nodes, self.y_node_], 'd_w3', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([self.y_node_]), name='d_b3', dtype=tf.float32)}\n",
    "            # Output shape has 2 features; Shape: [batch(real) + batch(gen.), 2]\n",
    "            \n",
    "        # Real data output\n",
    "        # Fully connect the layers\n",
    "        d1 = tf.nn.leaky_relu(tf.matmul(self.X_, d_h1['weights']) + d_h1['biases'])\n",
    "        if self.original_loss_fn_:\n",
    "            self.d2_ = tf.nn.leaky_relu(tf.matmul(d1, d_h2['weights']) + d_h2['biases']) # Originally acitvated with sigmoid\n",
    "        else:\n",
    "            self.d2_ = tf.matmul(d1, d_h2['weights']) + d_h2['biases']\n",
    "        d_out = tf.nn.sigmoid(tf.matmul(self.d2_, d_h3['weights']) + d_h3['biases'])\n",
    "        \n",
    "        self.y_data_ = d_out # 'y_data' == D(x)\n",
    "        \n",
    "        # Generated data output\n",
    "        # Fully connect the layers\n",
    "        gen_d1 = tf.nn.leaky_relu(tf.matmul(self.gen_X_, d_h1['weights']) + d_h1['biases'])\n",
    "        if self.original_loss_fn_:\n",
    "            self.gen_d2_ = tf.nn.leaky_relu(tf.matmul(gen_d1, d_h2['weights']) + d_h2['biases']) # Originally acitvated with sigmoid\n",
    "        else:\n",
    "            self.gen_d2_ = tf.matmul(gen_d1, d_h2['weights']) + d_h2['biases']\n",
    "        gen_d_out = tf.nn.sigmoid(tf.matmul(self.gen_d2_, d_h3['weights']) + d_h3['biases'])\n",
    "        \n",
    "        self.gen_y_ = gen_d_out # 'gen_y' == D[G(z)]\n",
    "        \n",
    "        ## List of 'var_list' for discriminator trainer to optimise ##\n",
    "        self.dis_params_ = [d_h1['weights'], d_h1['biases'],\n",
    "                           d_h2['weights'], d_h2['biases'],\n",
    "                           d_h3['weights'], d_h3['biases']]\n",
    "\n",
    "    def optimise(self, train_step=0.0001):\n",
    "        \n",
    "        if self.original_loss_fn_:\n",
    "            # Losses\n",
    "            # tf.reduce_mean() to compute the average losses\n",
    "            self.d_loss_ = - tf.reduce_mean(tf.math.log(self.y_data_) + tf.math.log(1.0 - self.gen_y_))\n",
    "#             self.g_loss_ = tf.reduce_mean(tf.math.log(1.0 - self.gen_y_))\n",
    "            self.g_loss_ = - tf.reduce_mean(tf.math.log(self.gen_y_))\n",
    "                # Alternate generator loss function optimisation: - tf.reduce_mean(tf.log(self.gen_y_))\n",
    "                \n",
    "        else:\n",
    "            # Alternate loss function: cross-entropy losses\n",
    "            D_loss_real = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=self.d2_, labels=tf.ones_like(self.d2_)))\n",
    "            D_loss_fake = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=self.gen_d2_, labels=tf.zeros_like(self.gen_d2_)))\n",
    "            self.d_loss_ = D_loss_real + D_loss_fake\n",
    "            self.g_loss_ = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=self.g2_, labels=tf.ones_like(self.g2_)))\n",
    " \n",
    "        # Optimiser\n",
    "        # optimiser = tf.train.AdadeltaOptimizer(learning_rate=1.0)\n",
    "        optimiser = tf.optimizers.Adam(learning_rate=train_step)\n",
    "        \n",
    "        # Optimisation Trainers\n",
    "#         self.d_trainer_ = optimiser.minimize(self.d_loss_, var_list=self.dis_params_)\n",
    "#         self.g_trainer_ = optimiser.minimize(self.g_loss_, var_list=self.gen_params_)     \n",
    "        self.d_trainer_ = tf.optimizers.SGD(learning_rate=train_step).minimize(self.d_loss_, var_list=self.dis_params_)\n",
    "        self.g_trainer_ = tf.optimizers.Adam(learning_rate=train_step).minimize(self.g_loss_, var_list=self.gen_params_)\n",
    "\n",
    "\n",
    "\n",
    "class WassersteinGAN(object):\n",
    "    \n",
    "    \"\"\" Invoke the methods in sequential order according to the scope, otherwise there would be errors. \"\"\"\n",
    "    \n",
    "    def __init__(self, batches=2048, X_nodes=29, y_nodes=1, z_dims=100, X_name='X_data', z_name='z_prior', orig_loss_fn=True):\n",
    "        \n",
    "        # Initialise inputs\n",
    "        self.batch_, self.X_node_, self.y_node_, self.z_dim_ = batches, X_nodes, y_nodes, z_dims\n",
    "        self.original_loss_fn_ = orig_loss_fn\n",
    "        \n",
    "        # Initialise placeholders\n",
    "        self.X_ = tf.compat.v1.placeholder(tf.float32, [None, self.X_node_], name=X_name)\n",
    "        self.prior_z_ = tf.compat.v1.placeholder(tf.float32, [None, self.z_dim_], name=z_name)\n",
    "        \n",
    "        # Generator parameters\n",
    "        self.gen_X_, self.gen_params_ = None, None\n",
    "        \n",
    "        # Discriminator parameters\n",
    "        self.y_data_, self.gen_y_, self.dis_params_ = None, None, None\n",
    "        \n",
    "        # Optimisation parameters\n",
    "        self.clip_dis_, self.d_loss_, self.g_loss_, self.d_trainer_, self.g_trainer_ = None, None, None, None, None\n",
    "        self.g2_, self.d2_, self.gen_d2_ = None, None, None\n",
    "                  \n",
    "    def generator(self, h1_nodes=500, h2_nodes=500):\n",
    "        \n",
    "        # Generator weights/biases\n",
    "        g_h1 = {'weights': tf.Variable(xavier_init([self.z_dim_, h1_nodes], 'g_w1', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros(h1_nodes), name='g_b1', dtype=tf.float32)}\n",
    "        g_h2 = {'weights': tf.Variable(xavier_init([h1_nodes, h2_nodes], 'g_w2', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([h2_nodes]), name='g_b2', dtype=tf.float32)}\n",
    "        g_h3 = {'weights': tf.Variable(xavier_init([h2_nodes, self.X_node_], 'g_w3', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([self.X_node_]), name='g_b3', dtype=tf.float32)}\n",
    "\n",
    "        # Fully connect the layers\n",
    "        g1 = tf.nn.leaky_relu(tf.matmul(self.prior_z_, g_h1['weights']) + g_h1['biases'])\n",
    "        if self.original_loss_fn_:\n",
    "            self.g2_ = tf.nn.leaky_relu(tf.matmul(g1, g_h2['weights']) + g_h2['biases'])\n",
    "        else:\n",
    "            self.g2_ = tf.matmul(g1, g_h2['weights']) + g_h2['biases']\n",
    "        g_out = tf.nn.tanh(tf.matmul(self.g2_, g_h3['weights']) + g_h3['biases'])\n",
    "        # Activation changed from sigmoid to tanh\n",
    "        \n",
    "        # Generate fake x's with from output layer of generator\n",
    "        self.gen_X_ = g_out\n",
    "\n",
    "        # List of 'var_list' for generator trainer to optimise\n",
    "        self.gen_params_ = [g_h1['weights'], g_h1['biases'],\n",
    "                           g_h2['weights'], g_h2['biases'],\n",
    "                           g_h3['weights'], g_h3['biases']]\n",
    "        \n",
    "    def discriminator(self, h1_nodes=500, h2_nodes=500):\n",
    "        \n",
    "        ## Discriminator weights/biases ##\n",
    "        d_h1 = {'weights': tf.Variable(xavier_init([self.X_node_, h1_nodes], 'd_w1', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([h1_nodes]), name='d_b1', dtype=tf.float32)}\n",
    "        d_h2 = {'weights': tf.Variable(xavier_init([h1_nodes, h2_nodes], 'd_w2', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([h2_nodes]), name='d_b2', dtype=tf.float32)}\n",
    "        d_h3 = {'weights': tf.Variable(xavier_init([h2_nodes, self.y_node_], 'd_w3', tf.float32)),\n",
    "                'biases': tf.Variable(tf.zeros([self.y_node_]), name='d_b3', dtype=tf.float32)}\n",
    "            # Output shape has 2 features; Shape: [batch(real) + batch(gen.), 2]\n",
    "            \n",
    "        # Real data output\n",
    "        # Fully connect the layers\n",
    "        d1 = tf.nn.leaky_relu(tf.matmul(self.X_, d_h1['weights']) + d_h1['biases'])\n",
    "        if self.original_loss_fn_:\n",
    "            self.d2_ = tf.nn.leaky_relu(tf.matmul(d1, d_h2['weights']) + d_h2['biases']) # Originally acitvated with sigmoid\n",
    "        else:\n",
    "            self.d2_ = tf.matmul(d1, d_h2['weights']) + d_h2['biases']\n",
    "        d_out = tf.nn.sigmoid(tf.matmul(self.d2_, d_h3['weights']) + d_h3['biases'])\n",
    "        \n",
    "        self.y_data_ = d_out # 'y_data' == D(x)\n",
    "        \n",
    "        # Generated data output\n",
    "        # Fully connect the layers\n",
    "        gen_d1 = tf.nn.leaky_relu(tf.matmul(self.gen_X_, d_h1['weights']) + d_h1['biases'])\n",
    "        if self.original_loss_fn_:\n",
    "            self.gen_d2_ = tf.nn.leaky_relu(tf.matmul(gen_d1, d_h2['weights']) + d_h2['biases']) # Originally acitvated with sigmoid\n",
    "        else:\n",
    "            self.gen_d2_ = tf.matmul(gen_d1, d_h2['weights']) + d_h2['biases']\n",
    "        gen_d_out = tf.nn.sigmoid(tf.matmul(self.gen_d2_, d_h3['weights']) + d_h3['biases'])\n",
    "        \n",
    "        self.gen_y_ = gen_d_out # 'gen_y' == D[G(z)]\n",
    "        \n",
    "        ## List of 'var_list' for discriminator trainer to optimise ##\n",
    "        self.dis_params_ = [d_h1['weights'], d_h1['biases'],\n",
    "                           d_h2['weights'], d_h2['biases'],\n",
    "                           d_h3['weights'], d_h3['biases']]\n",
    "        \n",
    "        # Clipping of discriminator\n",
    "        self.clip_dis_ = [param.assign(tf.clip_by_value(param, -0.05, 0.05)) for param in self.dis_params_]\n",
    "\n",
    "    def optimise(self, train_step=0.0001):\n",
    "        \n",
    "        if self.original_loss_fn_:\n",
    "            # Losses\n",
    "            # tf.reduce_mean() to compute the average losses\n",
    "            self.d_loss_ = - tf.reduce_mean(tf.math.log(self.y_data_) + tf.math.log(1.0 - self.gen_y_))\n",
    "#             self.g_loss_ = tf.reduce_mean(tf.math.log(1.0 - self.gen_y_))\n",
    "            self.g_loss_ = - tf.reduce_mean(tf.math.log(self.gen_y_))\n",
    "                # Alternate generator loss function optimisation to avoid vanishing gradients: - tf.reduce_mean(tf.log(self.gen_y_))\n",
    "        \n",
    "        else:     \n",
    "            # Alternate loss function: cross-entropy losses\n",
    "            D_loss_real = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=self.d2_, labels=tf.ones_like(self.d2_)))\n",
    "            D_loss_fake = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=self.gen_d2_, labels=tf.zeros_like(self.gen_d2_)))\n",
    "            self.d_loss_ = D_loss_real + D_loss_fake\n",
    "            self.g_loss_ = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=self.g2_, labels=tf.ones_like(self.g2_)))\n",
    " \n",
    "        # Optimiser\n",
    "        optimiser = tf.train.RMSPropOptimizer(learning_rate=train_step)\n",
    "        \n",
    "        # Optimisation Trainers\n",
    "#         self.d_trainer_ = optimiser.minimize(self.d_loss_, var_list=self.dis_params_)\n",
    "#         self.g_trainer_ = optimiser.minimize(self.g_loss_, var_list=self.gen_params_)\n",
    "        self.d_trainer_ = tf.optimizers.SGD(learning_rate=train_step)#.minimize(self.d_loss_, var_list=self.dis_params_)\n",
    "        self.g_trainer_ = tf.optimizers.Adam(learning_rate=train_step)#.minimize(self.g_loss_, var_list=self.gen_params_)\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3f434a51-290d-496f-af32-5662e4cddf55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time      0\n",
       "V1        0\n",
       "V2        0\n",
       "V3        0\n",
       "V4        0\n",
       "V5        0\n",
       "V6        0\n",
       "V7        0\n",
       "V8        0\n",
       "V9        0\n",
       "V10       0\n",
       "V11       0\n",
       "V12       0\n",
       "V13       0\n",
       "V14       0\n",
       "V15       0\n",
       "V16       0\n",
       "V17       0\n",
       "V18       0\n",
       "V19       0\n",
       "V20       0\n",
       "V21       0\n",
       "V22       0\n",
       "V23       0\n",
       "V24       0\n",
       "V25       0\n",
       "V26       0\n",
       "V27       0\n",
       "V28       0\n",
       "Amount    0\n",
       "Class     0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking for missing value\n",
    "credit.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "26e68911-d3d5-4192-8351-1ded85c9239e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Functions defined\n",
    "def uniform_draw_feat_class(data_df, target_name, draw_size):\n",
    "    \"\"\" Sample out 1 particular class out of the full data uniformly in a predefined sample size,\n",
    "    segregate pandas dataframe into their respective target classes. \"\"\"\n",
    "    \n",
    "    # Break down full dataset into their respective classes\n",
    "    target_class_dfs = {}\n",
    "    n_classes = data_df[target_name].unique()\n",
    "    \n",
    "    for unique_class in n_classes:\n",
    "        # Store each segregated class dataframes as dictionary\n",
    "        class_i_df = data_df[data_df[target_name] == unique_class]\n",
    "        class_i_df = (class_i_df.reset_index()).drop(columns=['index'])\n",
    "        target_class_dfs[unique_class] = class_i_df\n",
    "        \n",
    "    # Obtain uniform samples of size 'draw_size' and store as dictionary\n",
    "    uniform_class_sample_draws = {}\n",
    "    \n",
    "    # Pandas df. sampling default None equivalently uniform draws\n",
    "    for unique_class in n_classes:\n",
    "        unif_samples = target_class_dfs[unique_class].sample(n=draw_size, replace=True)\n",
    "        sample_df = pd.DataFrame(unif_samples, columns=data_df.columns)    \n",
    "        uniform_class_sample_draws[unique_class] = unif_samples\n",
    "    \n",
    "    # Returns dictionary of pandas dataframes, split into respective classes\n",
    "    return uniform_class_sample_draws\n",
    "\n",
    "def xavier_init(size, label, data_type):\n",
    "    \"\"\" For usage in tensorflow neural nets weights initialisation per layer.\n",
    "    Helps resolve issue of overfitting to data. \"\"\"\n",
    "    \n",
    "    tf.random.set_seed(42)\n",
    "    in_dim = size[0]; out_dim = size[1]\n",
    "    xavier_stddev = 1. / tf.sqrt((in_dim + out_dim) / 2.)\n",
    "    return tf.random.normal(shape=size, stddev=xavier_stddev, name=label, dtype=data_type)\n",
    "\n",
    "def steady_epoch(loss_lst, num_epochs_ran, sd_fluc_pct=0.15, scan_frame=225, stab_fn=min):   \n",
    "    \"\"\" \n",
    "    FOR THE PURPOSE OF FINDING A FRAME OF EPOCHS WHERE THE LOSSES HAS STABILISED WITHIN SPECIFIED PERCENTAGE OF \n",
    "    1SD OF LOSSES FOR ALL EPOCHS EXECUTED\n",
    "    \n",
    "    DEFAULT FRAME OF EPOCHS WHERE LOSS IS CONSIDERED STEADY IS 225, FLUCTUATING WITHIN +/- 15% OF 1 S.D. OF THE LOSSES\n",
    "    \n",
    "    [EXAMPLE - 5% OF S.D., 5% OF TOTAL EPOCHS RAN]\n",
    "    Obtain minimum of losses and its corresponding 1 s.d. of loss of the given array of losses\n",
    "    Do for a list of losses\n",
    "        For some 5% out of all epochs ran\n",
    "            if that particular 5% of epochs have losses each within +/-5% of the minimum loss,\n",
    "                Epoch frame chosen as steady epoch-period\n",
    "                Else return 'no steady epoch-periods found'; break\n",
    "        Return frame of epochs found \n",
    "    \"\"\"\n",
    "    # Obtain minimum and s.d. value of the losses for all epochs\n",
    "    loss_1sd = np.std(loss_lst)\n",
    "    stab_fn_loss = stab_fn(loss_lst)\n",
    "\n",
    "    epoch_frame_start = 0\n",
    "    epoch_frame_end = epoch_frame_start + scan_frame\n",
    "    \n",
    "    # Return nothing in the event no frame of losses are found to have steady losses\n",
    "    steady_frame=None\n",
    "    \n",
    "    # For loop termination\n",
    "    exit_loop = True\n",
    "    while exit_loop:\n",
    "\n",
    "        # Reset counter when one of the epoch is not within +/-'sd_fluc_pct'% of minimum losses\n",
    "        counter_5pct = 0\n",
    "\n",
    "        for epoch in range(epoch_frame_start, epoch_frame_end, 1):\n",
    "            if (loss_lst[epoch] > stab_fn_loss and loss_lst[epoch] <= (stab_fn_loss + sd_fluc_pct*loss_1sd)) \\\n",
    "            or (loss_lst[epoch] < stab_fn_loss and loss_lst[epoch] >= (stab_fn_loss - sd_fluc_pct*loss_1sd)):\n",
    "                \n",
    "                # Increase counter progressively until all (225) epochs in 'scan_frame' are within specified tolerance\n",
    "                counter_5pct += 1\n",
    "\n",
    "                # When (225) epochs are within +/-'sd_fluc_pct'% of minimal loss, flag the (225th) final epoch\n",
    "                if counter_5pct == scan_frame:\n",
    "                    print('Steady epoch frame found at epoch {} as final'.format(epoch_frame_end))\n",
    "                    \n",
    "                    # Return numpy array of epoch numbers within (225) specified range\n",
    "                    steady_frame = np.linspace(start=(epoch_frame_end-scan_frame+1), stop=epoch_frame_end, num=scan_frame)\n",
    "                    exit_loop = False\n",
    "                    break           \n",
    "            else:\n",
    "                break\n",
    "\n",
    "        # When a frame of 225 epochs within +/-'sd_fluc_pct'% of minimum losses is not found,\n",
    "            # shift to next frame by 1 epoch\n",
    "        epoch_frame_start += 1\n",
    "        epoch_frame_end += 1\n",
    "\n",
    "        if epoch_frame_end > num_epochs_ran:\n",
    "            print('No steady epoch frame found!')\n",
    "            break\n",
    "\n",
    "    return steady_frame\n",
    "     \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "cfa998d0-1d81-4146-9d98-bb8593e96eba",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "### Load credit card data and preprocessing ###\n",
    "\n",
    "\n",
    "# Convert data into 'mini-batchable' numpy NORMAL arrays, drop 'Time'\n",
    "credit_X = credit.drop(labels=['Time','Class'], axis='columns')\n",
    "credit_y = credit['Class'].to_numpy()\n",
    "\n",
    "# Train-test split data: 75/25\n",
    "X_train, X_test, y_train, y_test = train_test_split(credit_X, credit_y, test_size=0.25, random_state=seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "97019717-ec0e-437e-baa9-3321e2f8df66",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 284807 entries, 0 to 284806\n",
      "Data columns (total 31 columns):\n",
      " #   Column  Dtype  \n",
      "---  ------  -----  \n",
      " 0   Time    float64\n",
      " 1   V1      float64\n",
      " 2   V2      float64\n",
      " 3   V3      float64\n",
      " 4   V4      float64\n",
      " 5   V5      float64\n",
      " 6   V6      float64\n",
      " 7   V7      float64\n",
      " 8   V8      float64\n",
      " 9   V9      float64\n",
      " 10  V10     float64\n",
      " 11  V11     float64\n",
      " 12  V12     float64\n",
      " 13  V13     float64\n",
      " 14  V14     float64\n",
      " 15  V15     float64\n",
      " 16  V16     float64\n",
      " 17  V17     float64\n",
      " 18  V18     float64\n",
      " 19  V19     float64\n",
      " 20  V20     float64\n",
      " 21  V21     float64\n",
      " 22  V22     float64\n",
      " 23  V23     float64\n",
      " 24  V24     float64\n",
      " 25  V25     float64\n",
      " 26  V26     float64\n",
      " 27  V27     float64\n",
      " 28  V28     float64\n",
      " 29  Amount  float64\n",
      " 30  Class   int64  \n",
      "dtypes: float64(30), int64(1)\n",
      "memory usage: 67.4 MB\n",
      "None \n",
      "\n",
      "                Time            V1            V2            V3            V4  \\\n",
      "count  284807.000000  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
      "mean    94813.859575  1.168375e-15  3.416908e-16 -1.379537e-15  2.074095e-15   \n",
      "std     47488.145955  1.958696e+00  1.651309e+00  1.516255e+00  1.415869e+00   \n",
      "min         0.000000 -5.640751e+01 -7.271573e+01 -4.832559e+01 -5.683171e+00   \n",
      "25%     54201.500000 -9.203734e-01 -5.985499e-01 -8.903648e-01 -8.486401e-01   \n",
      "50%     84692.000000  1.810880e-02  6.548556e-02  1.798463e-01 -1.984653e-02   \n",
      "75%    139320.500000  1.315642e+00  8.037239e-01  1.027196e+00  7.433413e-01   \n",
      "max    172792.000000  2.454930e+00  2.205773e+01  9.382558e+00  1.687534e+01   \n",
      "\n",
      "                 V5            V6            V7            V8            V9  \\\n",
      "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
      "mean   9.604066e-16  1.487313e-15 -5.556467e-16  1.213481e-16 -2.406331e-15   \n",
      "std    1.380247e+00  1.332271e+00  1.237094e+00  1.194353e+00  1.098632e+00   \n",
      "min   -1.137433e+02 -2.616051e+01 -4.355724e+01 -7.321672e+01 -1.343407e+01   \n",
      "25%   -6.915971e-01 -7.682956e-01 -5.540759e-01 -2.086297e-01 -6.430976e-01   \n",
      "50%   -5.433583e-02 -2.741871e-01  4.010308e-02  2.235804e-02 -5.142873e-02   \n",
      "75%    6.119264e-01  3.985649e-01  5.704361e-01  3.273459e-01  5.971390e-01   \n",
      "max    3.480167e+01  7.330163e+01  1.205895e+02  2.000721e+01  1.559499e+01   \n",
      "\n",
      "                V10           V11           V12           V13           V14  \\\n",
      "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
      "mean   2.239053e-15  1.673327e-15 -1.247012e-15  8.190001e-16  1.207294e-15   \n",
      "std    1.088850e+00  1.020713e+00  9.992014e-01  9.952742e-01  9.585956e-01   \n",
      "min   -2.458826e+01 -4.797473e+00 -1.868371e+01 -5.791881e+00 -1.921433e+01   \n",
      "25%   -5.354257e-01 -7.624942e-01 -4.055715e-01 -6.485393e-01 -4.255740e-01   \n",
      "50%   -9.291738e-02 -3.275735e-02  1.400326e-01 -1.356806e-02  5.060132e-02   \n",
      "75%    4.539234e-01  7.395934e-01  6.182380e-01  6.625050e-01  4.931498e-01   \n",
      "max    2.374514e+01  1.201891e+01  7.848392e+00  7.126883e+00  1.052677e+01   \n",
      "\n",
      "                V15           V16           V17           V18           V19  \\\n",
      "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
      "mean   4.887456e-15  1.437716e-15 -3.772171e-16  9.564149e-16  1.039917e-15   \n",
      "std    9.153160e-01  8.762529e-01  8.493371e-01  8.381762e-01  8.140405e-01   \n",
      "min   -4.498945e+00 -1.412985e+01 -2.516280e+01 -9.498746e+00 -7.213527e+00   \n",
      "25%   -5.828843e-01 -4.680368e-01 -4.837483e-01 -4.988498e-01 -4.562989e-01   \n",
      "50%    4.807155e-02  6.641332e-02 -6.567575e-02 -3.636312e-03  3.734823e-03   \n",
      "75%    6.488208e-01  5.232963e-01  3.996750e-01  5.008067e-01  4.589494e-01   \n",
      "max    8.877742e+00  1.731511e+01  9.253526e+00  5.041069e+00  5.591971e+00   \n",
      "\n",
      "                V20           V21           V22           V23           V24  \\\n",
      "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05   \n",
      "mean   6.406204e-16  1.654067e-16 -3.568593e-16  2.578648e-16  4.473266e-15   \n",
      "std    7.709250e-01  7.345240e-01  7.257016e-01  6.244603e-01  6.056471e-01   \n",
      "min   -5.449772e+01 -3.483038e+01 -1.093314e+01 -4.480774e+01 -2.836627e+00   \n",
      "25%   -2.117214e-01 -2.283949e-01 -5.423504e-01 -1.618463e-01 -3.545861e-01   \n",
      "50%   -6.248109e-02 -2.945017e-02  6.781943e-03 -1.119293e-02  4.097606e-02   \n",
      "75%    1.330408e-01  1.863772e-01  5.285536e-01  1.476421e-01  4.395266e-01   \n",
      "max    3.942090e+01  2.720284e+01  1.050309e+01  2.252841e+01  4.584549e+00   \n",
      "\n",
      "                V25           V26           V27           V28         Amount  \\\n",
      "count  2.848070e+05  2.848070e+05  2.848070e+05  2.848070e+05  284807.000000   \n",
      "mean   5.340915e-16  1.683437e-15 -3.660091e-16 -1.227390e-16      88.349619   \n",
      "std    5.212781e-01  4.822270e-01  4.036325e-01  3.300833e-01     250.120109   \n",
      "min   -1.029540e+01 -2.604551e+00 -2.256568e+01 -1.543008e+01       0.000000   \n",
      "25%   -3.171451e-01 -3.269839e-01 -7.083953e-02 -5.295979e-02       5.600000   \n",
      "50%    1.659350e-02 -5.213911e-02  1.342146e-03  1.124383e-02      22.000000   \n",
      "75%    3.507156e-01  2.409522e-01  9.104512e-02  7.827995e-02      77.165000   \n",
      "max    7.519589e+00  3.517346e+00  3.161220e+01  3.384781e+01   25691.160000   \n",
      "\n",
      "               Class  \n",
      "count  284807.000000  \n",
      "mean        0.001727  \n",
      "std         0.041527  \n",
      "min         0.000000  \n",
      "25%         0.000000  \n",
      "50%         0.000000  \n",
      "75%         0.000000  \n",
      "max         1.000000   \n",
      "\n",
      "Class\n",
      "0    284315\n",
      "1       492\n",
      "Name: count, dtype: int64 \n",
      "\n"
     ]
    },
    {
     "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>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>V10</th>\n",
       "      <th>V11</th>\n",
       "      <th>V12</th>\n",
       "      <th>V13</th>\n",
       "      <th>V14</th>\n",
       "      <th>V15</th>\n",
       "      <th>V16</th>\n",
       "      <th>V17</th>\n",
       "      <th>V18</th>\n",
       "      <th>V19</th>\n",
       "      <th>V20</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>0.090794</td>\n",
       "      <td>-0.551600</td>\n",
       "      <td>-0.617801</td>\n",
       "      <td>-0.991390</td>\n",
       "      <td>-0.311169</td>\n",
       "      <td>1.468177</td>\n",
       "      <td>-0.470401</td>\n",
       "      <td>0.207971</td>\n",
       "      <td>0.025791</td>\n",
       "      <td>0.403993</td>\n",
       "      <td>0.251412</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>-0.166974</td>\n",
       "      <td>1.612727</td>\n",
       "      <td>1.065235</td>\n",
       "      <td>0.489095</td>\n",
       "      <td>-0.143772</td>\n",
       "      <td>0.635558</td>\n",
       "      <td>0.463917</td>\n",
       "      <td>-0.114805</td>\n",
       "      <td>-0.183361</td>\n",
       "      <td>-0.145783</td>\n",
       "      <td>-0.069083</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>0.207643</td>\n",
       "      <td>0.624501</td>\n",
       "      <td>0.066084</td>\n",
       "      <td>0.717293</td>\n",
       "      <td>-0.165946</td>\n",
       "      <td>2.345865</td>\n",
       "      <td>-2.890083</td>\n",
       "      <td>1.109969</td>\n",
       "      <td>-0.121359</td>\n",
       "      <td>-2.261857</td>\n",
       "      <td>0.524980</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>-0.054952</td>\n",
       "      <td>-0.226487</td>\n",
       "      <td>0.178228</td>\n",
       "      <td>0.507757</td>\n",
       "      <td>-0.287924</td>\n",
       "      <td>-0.631418</td>\n",
       "      <td>-1.059647</td>\n",
       "      <td>-0.684093</td>\n",
       "      <td>1.965775</td>\n",
       "      <td>-1.232622</td>\n",
       "      <td>-0.208038</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>0.753074</td>\n",
       "      <td>-0.822843</td>\n",
       "      <td>0.538196</td>\n",
       "      <td>1.345852</td>\n",
       "      <td>-1.119670</td>\n",
       "      <td>0.175121</td>\n",
       "      <td>-0.451449</td>\n",
       "      <td>-0.237033</td>\n",
       "      <td>-0.038195</td>\n",
       "      <td>0.803487</td>\n",
       "      <td>0.408542</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9       V10       V11       V12       V13       V14  \\\n",
       "0  0.098698  0.363787  0.090794 -0.551600 -0.617801 -0.991390 -0.311169   \n",
       "1  0.085102 -0.255425 -0.166974  1.612727  1.065235  0.489095 -0.143772   \n",
       "2  0.247676 -1.514654  0.207643  0.624501  0.066084  0.717293 -0.165946   \n",
       "3  0.377436 -1.387024 -0.054952 -0.226487  0.178228  0.507757 -0.287924   \n",
       "4 -0.270533  0.817739  0.753074 -0.822843  0.538196  1.345852 -1.119670   \n",
       "\n",
       "        V15       V16       V17       V18       V19       V20       V21  \\\n",
       "0  1.468177 -0.470401  0.207971  0.025791  0.403993  0.251412 -0.018307   \n",
       "1  0.635558  0.463917 -0.114805 -0.183361 -0.145783 -0.069083 -0.225775   \n",
       "2  2.345865 -2.890083  1.109969 -0.121359 -2.261857  0.524980  0.247998   \n",
       "3 -0.631418 -1.059647 -0.684093  1.965775 -1.232622 -0.208038 -0.108300   \n",
       "4  0.175121 -0.451449 -0.237033 -0.038195  0.803487  0.408542 -0.009431   \n",
       "\n",
       "        V22       V23       V24       V25       V26       V27       V28  \\\n",
       "0  0.277838 -0.110474  0.066928  0.128539 -0.189115  0.133558 -0.021053   \n",
       "1 -0.638672  0.101288 -0.339846  0.167170  0.125895 -0.008983  0.014724   \n",
       "2  0.771679  0.909412 -0.689281 -0.327642 -0.139097 -0.055353 -0.059752   \n",
       "3  0.005274 -0.190321 -1.175575  0.647376 -0.221929  0.062723  0.061458   \n",
       "4  0.798278 -0.137458  0.141267 -0.206010  0.502292  0.219422  0.215153   \n",
       "\n",
       "   Amount  Class  \n",
       "0  149.62      0  \n",
       "1    2.69      0  \n",
       "2  378.66      0  \n",
       "3  123.50      0  \n",
       "4   69.99      0  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "### Overview of data in 'credit' ###\n",
    "\n",
    "pd.set_option('display.max_columns', 1000)\n",
    "pd.set_option('display.max_rows', 1000)\n",
    "pd.set_option('max_info_columns', 1000)\n",
    "pd.set_option('max_info_rows', 1000)\n",
    "\n",
    "# Dtype; checking for nulls\n",
    "print(credit.info(), '\\n')\n",
    "\n",
    "# Descriptive statistics\n",
    "print(credit.describe(include='all'), '\\n')\n",
    "\n",
    "# Validate only fraud and non-fraud classes\n",
    "print(credit['Class'].value_counts(), '\\n')\n",
    "\n",
    "# 1st 5 rows\n",
    "credit.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c26436c7-7f3c-4a67-8725-24d67126f17f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total \"Amount\" in training set: 42513.46\n"
     ]
    }
   ],
   "source": [
    "# Train test split as how data was done earlier\n",
    "X_amt_tr, X_amt_te, y_amt_tr, y_amt_te = train_test_split(credit.drop(columns=['Class']).to_numpy(), credit['Class'].to_numpy(), \n",
    "                                                          test_size=0.25, random_state=42)\n",
    "\n",
    "# Create dataframes\n",
    "amt_tr = np.concatenate((X_amt_tr, y_amt_tr.reshape(-1, 1)), axis=1)\n",
    "amt_tr_df = pd.DataFrame(amt_tr, columns=credit.columns)\n",
    "amt_te = np.concatenate((X_amt_te, y_amt_te.reshape(-1, 1)), axis=1)\n",
    "amt_te_df = pd.DataFrame(amt_te, columns=credit.columns)\n",
    "\n",
    "# Training Set Fraud 'Amount' Sum\n",
    "tl_tr_fraud_amt = amt_tr_df[amt_tr_df['Class'] == 1]['Amount'].sum()\n",
    "# Test Set Fraud 'Amount' Sum\n",
    "tl_te_fraud_amt = amt_te_df[amt_te_df['Class'] == 1]['Amount'].sum()\n",
    "print('Total \"Amount\" in training set: {}'.format(np.round(tl_tr_fraud_amt, 2),np.round(tl_te_fraud_amt, 2)))\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b24317d1-edb2-4a17-ad49-b3669dee65e5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of frauds in training data: 379 out of 213605 cases (0.1774303036% fraud)\n",
      "Number of frauds in test data: 113 out of 71202 cases (0.1587034072% fraud)\n"
     ]
    }
   ],
   "source": [
    "# Labels\n",
    "X_cols = list(credit.drop(labels=['Time', 'Class'], axis='columns').columns)\n",
    "y_cols = credit[['Class']].columns\n",
    "target_names = credit['Class'].map(lambda x: 'Non-Fraud' if x==0 else 'Fraud').unique()\n",
    "\n",
    "# Training data as dataframe\n",
    "X_train_df = pd.DataFrame(X_train, columns=X_cols)\n",
    "X_train_df['Amount'] = StandardScaler().fit_transform(X_train_df['Amount'].to_numpy().reshape(-1, 1))\n",
    "X_train_df = X_train_df.reset_index().drop(columns=['index'])\n",
    "y_train_df = pd.DataFrame(y_train, columns=y_cols)\n",
    "# Full dataframe\n",
    "train_df = pd.concat([X_train_df, y_train_df], join='outer', axis='columns')\n",
    "# X, y as numpy array\n",
    "X_train, y_train = X_train_df.to_numpy(), y_train_df.to_numpy()\n",
    "\n",
    "# Test data as dataframe\n",
    "X_test_df = pd.DataFrame(X_test, columns=X_cols)\n",
    "X_test_df['Amount'] = StandardScaler().fit(X_train_df['Amount'].to_numpy().reshape(-1, 1)).transform(X_test_df['Amount'].to_numpy().reshape(-1, 1))\n",
    "X_test_df = X_test_df.reset_index().drop(columns=['index'])\n",
    "y_test_df = pd.DataFrame(y_test, columns=y_cols)\n",
    "# Full dataframe\n",
    "test_df = pd.concat([X_test_df, y_test_df], join='outer', axis='columns')\n",
    "# X, y as numpy array\n",
    "X_test, y_test = X_test_df.to_numpy(), y_test_df.to_numpy()\n",
    "\n",
    "# Check number of frauds in both training and testing data\n",
    "print('Number of frauds in training data: {} out of {} cases ({:.10f}% fraud)'.format(np.sum(y_train), y_train.shape[0], np.mean(y_train)*100))\n",
    "print('Number of frauds in test data: {} out of {} cases ({:.10f}% fraud)'.format(np.sum(y_test), y_test.shape[0], np.mean(y_test)*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c6427d39-440d-47fd-a2f4-5afd254a8a9d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "explained variance ratio (first two components): [0.12503723 0.10004168]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 489,
       "width": 590
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# 2-component PCA plot\n",
    "    # PCA is sensitive to dimensional scales, thus the need to standardise matrix X\n",
    "X_pca = copy.deepcopy(X_train)\n",
    "y_pca = copy.deepcopy(y_train).reshape(-1, )\n",
    "\n",
    "pca = PCA(n_components=2, random_state=seed)\n",
    "X_r = pca.fit_transform(X_pca)\n",
    "\n",
    "# Percentage of variance explained for each components\n",
    "print('explained variance ratio (first two components): %s'\n",
    "      % str(pca.explained_variance_ratio_))\n",
    "\n",
    "plt.figure(figsize=(6,5)); colors = ['brown', 'cyan']; lw = 2\n",
    "\n",
    "for color, i, target_name in zip(colors, [0, 1], target_names):\n",
    "    plt.scatter(X_r[y_pca == i, 0], X_r[y_pca == i, 1], color=color, alpha=.6, lw=lw, label=target_name)\n",
    "plt.legend(loc='best', shadow=False, scatterpoints=1)\n",
    "plt.title('PCA of Fraud/Non-Fraud')\n",
    "plt.tight_layout(); plt.margins(0.02)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "fd7754f9-41f7-47a6-b881-149d8b941611",
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Kernel Density Distribution subplots against target - 'Class'\n",
    "\n",
    "# # Scatter subplots\n",
    "# plt.figure(figsize=(20,18))\n",
    "# for plot, feat in enumerate(X_cols):\n",
    "    \n",
    "#     plt.subplot(5, 6, (plot+1))\n",
    "#     title = 'Fraud/Non-Fraud & ' + feat\n",
    "    \n",
    "#     # Normalise to visualise the differences in distributions\n",
    "#     temp_df = pd.concat([X_train_df[[feat]], y_train_df], axis='columns')\n",
    "#     temp_df.groupby(by='Class')[feat].plot(kind='kde', alpha=0.7, legend='best', lw=2.5)\n",
    "#     plt.title(title)\n",
    "#     plt.tight_layout(); plt.margins(0.02)\n",
    "    \n",
    "#     plt.show()\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "33be9f63-aa14-4540-8c84-0dcb0b00bff5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature \"V13\" failed to be rejected at 5% level with p-value 0.0973465482\n",
      "Feature \"V15\" failed to be rejected at 5% level with p-value 0.8092477538\n",
      "Feature \"V22\" failed to be rejected at 5% level with p-value 0.3224055980\n",
      "Feature \"V23\" failed to be rejected at 5% level with p-value 0.7011952304\n",
      "Feature \"V26\" failed to be rejected at 5% level with p-value 0.5115929524\n",
      "Feature \"Amount\" failed to be rejected at 5% level with p-value 0.9531649675\n",
      "\n",
      " Wilcoxon Rank-Sum Relevant Features:  ['V1', 'V2', 'V3', 'V4', 'V5', 'V6', 'V7', 'V8', 'V9', 'V10', 'V11', 'V12', 'V14', 'V16', 'V17', 'V18', 'V19', 'V20', 'V21', 'V24', 'V25', 'V27', 'V28']\n",
      "Total number of features selected: 23\n"
     ]
    }
   ],
   "source": [
    "# Perform Wilcoxon Rank-Sum Test at 5% Level to determine feature relevance\n",
    "\n",
    "# Keep results if needed later\n",
    "wilcox_result = dict([])\n",
    "\n",
    "for feat in X_cols:   \n",
    "    temp_df = pd.concat([X_train_df[[feat]], y_train_df], axis='columns')\n",
    "    \n",
    "    # Draw 300 samples (only 379 fraud data in training set, attempt to draw close to the hilt)\n",
    "    random.seed(42)\n",
    "    fraud_wilcox = random.sample(list(temp_df[temp_df['Class']==1][feat]), k=300)\n",
    "    nonfraud_wilcox = random.sample(list(temp_df[temp_df['Class']==0][feat]), k=300)\n",
    "    \n",
    "    rank_sums, p_val_wilcox = stats.wilcoxon(fraud_wilcox, nonfraud_wilcox)\n",
    "    \n",
    "    if p_val_wilcox > 0.05:\n",
    "        wilcox_result[feat] = ('No_Diff', p_val_wilcox)\n",
    "        print('Feature \"{}\" failed to be rejected at 5% level with p-value {:.10f}'.format(feat, p_val_wilcox))\n",
    "    else:\n",
    "        wilcox_result[feat] = ('Diff', p_val_wilcox)\n",
    "    \n",
    "# Rank-Sum Test determined differences in classes within feature\n",
    "wilcox_feat = [feat for feat in wilcox_result.keys() if wilcox_result[feat][0] == 'Diff']\n",
    "print('\\n', 'Wilcoxon Rank-Sum Relevant Features: ', wilcox_feat)\n",
    "print('Total number of features selected: {}'.format(len(wilcox_feat)))\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9501ade1-0fb5-47d4-b1bb-5b189dffb773",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unselected features: ['V26', 'V15', 'Amount', 'V13', 'V23', 'V22']\n"
     ]
    }
   ],
   "source": [
    "unselected_feats = list(set(X_cols).difference(wilcox_feat))\n",
    "print('Unselected features: {}'.format(unselected_feats))\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "da038172-a707-41d7-b503-e16b199fa982",
   "metadata": {},
   "outputs": [
    {
     "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></th>\n",
       "      <th>pearson_rho</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>V7</th>\n",
       "      <th>V5</th>\n",
       "      <td>0.030143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <td>0.016899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V7</th>\n",
       "      <th>V3</th>\n",
       "      <td>0.016640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <td>0.016007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <td>0.015015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>V7</th>\n",
       "      <th>V1</th>\n",
       "      <td>0.012172</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       pearson_rho\n",
       "V7 V5     0.030143\n",
       "V5 V6     0.016899\n",
       "V7 V3     0.016640\n",
       "V6 V7     0.016007\n",
       "V1 V2     0.015015\n",
       "V7 V1     0.012172"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# Obtain the top 6 most correlated wilcoxon-selected variables' to observe their scatter pair plot relations\n",
    "\n",
    "# Check for high correlations\n",
    "wilcox_corr = X_train_df.drop(columns=['V15', 'Amount', 'V13', 'V26', 'V22', 'V23']).corr().abs()\n",
    "wilcox_corr_pdSr = wilcox_corr.unstack()\n",
    "wilcox_corr_pdSr_sort = wilcox_corr_pdSr.sort_values(ascending=False).to_frame()\n",
    "\n",
    "# Top 6 correlated variable pairs\n",
    "wilcox_corr_pdSr_sort.iloc[23:35:2, :].rename(index=str, columns={0:'pearson_rho'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "49ca201c-5fc2-4323-9c02-a6e113a88b38",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "09f86d81-2130-4d35-b9fd-c7ac86200361",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Time elapsed to train:  7.923961877822876\n",
      "Prediction Score:  0.9646498693856914 \n",
      "\n",
      "Confusion Matrix:  \n",
      " [[68580  2509]\n",
      " [    8   105]] \n",
      "\n",
      "Classification Report: \n",
      "                    pre       rec       spe        f1       geo       iba       sup\n",
      "\n",
      "          0       1.00      0.96      0.93      0.98      0.95      0.90     71089\n",
      "          1       0.04      0.93      0.96      0.08      0.95      0.89       113\n",
      "\n",
      "avg / total       1.00      0.96      0.93      0.98      0.95      0.90     71202\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Using default values except \"penalty='elasticnet'\"\n",
    "# To preserve/remove meaningful/irrelevant Zfeatures without absolute purging\n",
    "sgd_clf = SGDClassifier(penalty='elasticnet', random_state=seed, class_weight='balanced')\n",
    "\n",
    "# Measure Start Time\n",
    "start_time = time.time()\n",
    "\n",
    "# Fit and obtain predictions\n",
    "sgd_clf.fit(X_train_df.drop(columns=['V15', 'V13', 'V23', 'Amount', 'V26', 'V22']).to_numpy(), y_train)\n",
    "y_hat = sgd_clf.predict(X_test_df.drop(columns=['V15', 'V13', 'V23', 'Amount', 'V26', 'V22']).to_numpy())\n",
    "\n",
    "# Measure End Time\n",
    "elapsed_time = time.time() - start_time\n",
    "print('Time elapsed to train: ', elapsed_time)\n",
    "\n",
    "pred_score = sgd_clf.score(X_test_df.drop(columns=['V15', 'V13', 'V23', 'Amount', 'V26', 'V22']).to_numpy(), y_test)\n",
    "print('Prediction Score: ', pred_score, '\\n')\n",
    "print('Confusion Matrix: ', '\\n', confusion_matrix(y_test, y_hat), '\\n')\n",
    "print('Classification Report:', '\\n', classification_report_imbalanced(y_test, y_hat))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "eb06f681-89b8-447a-b798-d02482994bc4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# General Parameters\n",
    "\n",
    "# Inputs\n",
    "gan_epochs = 6300\n",
    "mini_bat = 5000 # Equivalently gan_fraudless/fraud.batch_\n",
    "\n",
    "X_feats = len(wilcox_feat) # 30 features originally describing fraud/non-fraud; Equivalently gan_fraudless/fraud.X_node_\n",
    "prior_z_dim = 100 # Draw 100 prior values to feed generator; Equivalently gan_fraudless/fraud.z_dim_\n",
    "\n",
    "# Output\n",
    "y_outputs = 1 # Equivalently gan_fraudless/fraud.y_node_\n",
    "\n",
    "# Create GAN objects for fraud data\n",
    "gan_fraud = GAN(batches=mini_bat, X_nodes=X_feats, y_nodes=y_outputs, z_dims=prior_z_dim)\n",
    "gan_fraud.generator()\n",
    "gan_fraud.discriminator()\n",
    "gan_fraud.optimise()\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "19415fb3-2a8a-4cee-8713-e052a9bd1f5b",
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "Argument `fetch` = <keras.src.optimizers.sgd.SGD object at 0x000001AD1B09E5D0> has invalid type \"SGD\" must be a string or Tensor. (Can not convert a SGD into a Tensor or Operation.)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:308\u001b[0m, in \u001b[0;36m_ElementFetchMapper.__init__\u001b[1;34m(self, fetches, contraction_fn)\u001b[0m\n\u001b[0;32m    307\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 308\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_unique_fetches\u001b[38;5;241m.\u001b[39mappend(\u001b[43mops\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_default_graph\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mas_graph_element\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m    309\u001b[0m \u001b[43m      \u001b[49m\u001b[43mfetch\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mallow_tensor\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mallow_operation\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m)\n\u001b[0;32m    310\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\framework\\ops.py:2851\u001b[0m, in \u001b[0;36mGraph.as_graph_element\u001b[1;34m(self, obj, allow_tensor, allow_operation)\u001b[0m\n\u001b[0;32m   2850\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_lock:\n\u001b[1;32m-> 2851\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_as_graph_element_locked\u001b[49m\u001b[43m(\u001b[49m\u001b[43mobj\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mallow_tensor\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mallow_operation\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\framework\\ops.py:2947\u001b[0m, in \u001b[0;36mGraph._as_graph_element_locked\u001b[1;34m(self, obj, allow_tensor, allow_operation)\u001b[0m\n\u001b[0;32m   2945\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m   2946\u001b[0m   \u001b[38;5;66;03m# We give up!\u001b[39;00m\n\u001b[1;32m-> 2947\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCan not convert a \u001b[39m\u001b[38;5;132;01m%s\u001b[39;00m\u001b[38;5;124m into a \u001b[39m\u001b[38;5;132;01m%s\u001b[39;00m\u001b[38;5;124m.\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;241m%\u001b[39m\n\u001b[0;32m   2948\u001b[0m                   (\u001b[38;5;28mtype\u001b[39m(obj)\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m, types_str))\n",
      "\u001b[1;31mTypeError\u001b[0m: Can not convert a SGD into a Tensor or Operation.",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[62], line 34\u001b[0m\n\u001b[0;32m     31\u001b[0m fraud_z_sample \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39mrandom\u001b[38;5;241m.\u001b[39mnormal(size\u001b[38;5;241m=\u001b[39m[mini_bat, prior_z_dim])\u001b[38;5;241m.\u001b[39mastype(np\u001b[38;5;241m.\u001b[39mfloat32)\n\u001b[0;32m     33\u001b[0m \u001b[38;5;66;03m# Minimise Losses\u001b[39;00m\n\u001b[1;32m---> 34\u001b[0m _, fraud_dloss \u001b[38;5;241m=\u001b[39m \u001b[43msess\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43m[\u001b[49m\u001b[43mgan_fraud\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43md_trainer_\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgan_fraud\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43md_loss_\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[0;32m     35\u001b[0m \u001b[43m                              \u001b[49m\u001b[43mfeed_dict\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43m{\u001b[49m\u001b[43mgan_fraud\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mX_\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43mfraud_X_mb\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\n\u001b[0;32m     36\u001b[0m \u001b[43m                                         \u001b[49m\u001b[43mgan_fraud\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mprior_z_\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43mfraud_z_sample\u001b[49m\u001b[43m}\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     37\u001b[0m _, fraud_gloss \u001b[38;5;241m=\u001b[39m sess\u001b[38;5;241m.\u001b[39mrun([gan_fraud\u001b[38;5;241m.\u001b[39mg_trainer_, gan_fraud\u001b[38;5;241m.\u001b[39mg_loss_], \n\u001b[0;32m     38\u001b[0m                               feed_dict\u001b[38;5;241m=\u001b[39m{gan_fraud\u001b[38;5;241m.\u001b[39mX_: fraud_X_mb,\n\u001b[0;32m     39\u001b[0m                                          gan_fraud\u001b[38;5;241m.\u001b[39mprior_z_: fraud_z_sample})\n\u001b[0;32m     41\u001b[0m \u001b[38;5;66;03m# Store losses per epoch\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:972\u001b[0m, in \u001b[0;36mBaseSession.run\u001b[1;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m    969\u001b[0m run_metadata_ptr \u001b[38;5;241m=\u001b[39m tf_session\u001b[38;5;241m.\u001b[39mTF_NewBuffer() \u001b[38;5;28;01mif\u001b[39;00m run_metadata \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m    971\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 972\u001b[0m   result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfetches\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfeed_dict\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43moptions_ptr\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    973\u001b[0m \u001b[43m                     \u001b[49m\u001b[43mrun_metadata_ptr\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    974\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m run_metadata:\n\u001b[0;32m    975\u001b[0m     proto_data \u001b[38;5;241m=\u001b[39m tf_session\u001b[38;5;241m.\u001b[39mTF_GetBuffer(run_metadata_ptr)\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:1200\u001b[0m, in \u001b[0;36mBaseSession._run\u001b[1;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[0;32m   1197\u001b[0m       feed_map[compat\u001b[38;5;241m.\u001b[39mas_bytes(subfeed_t\u001b[38;5;241m.\u001b[39mname)] \u001b[38;5;241m=\u001b[39m (subfeed_t, subfeed_val)\n\u001b[0;32m   1199\u001b[0m \u001b[38;5;66;03m# Create a fetch handler to take care of the structure of fetches.\u001b[39;00m\n\u001b[1;32m-> 1200\u001b[0m fetch_handler \u001b[38;5;241m=\u001b[39m \u001b[43m_FetchHandler\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m   1201\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfetches\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfeed_dict_tensor\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfeed_handles\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mfeed_handles\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m   1203\u001b[0m \u001b[38;5;66;03m# Run request and get response.\u001b[39;00m\n\u001b[0;32m   1204\u001b[0m \u001b[38;5;66;03m# We need to keep the returned movers alive for the following _do_run().\u001b[39;00m\n\u001b[0;32m   1205\u001b[0m \u001b[38;5;66;03m# These movers are no longer needed when _do_run() completes, and\u001b[39;00m\n\u001b[0;32m   1206\u001b[0m \u001b[38;5;66;03m# are deleted when `movers` goes out of scope when this _run() ends.\u001b[39;00m\n\u001b[0;32m   1207\u001b[0m \u001b[38;5;66;03m# TODO(yuanbyu, keveman): Revisit whether we should just treat feeding\u001b[39;00m\n\u001b[0;32m   1208\u001b[0m \u001b[38;5;66;03m# of a handle from a different device as an error.\u001b[39;00m\n\u001b[0;32m   1209\u001b[0m _ \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_update_with_movers(feed_dict_tensor, feed_map)\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:489\u001b[0m, in \u001b[0;36m_FetchHandler.__init__\u001b[1;34m(self, graph, fetches, feeds, feed_handles)\u001b[0m\n\u001b[0;32m    477\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Creates a fetch handler.\u001b[39;00m\n\u001b[0;32m    478\u001b[0m \n\u001b[0;32m    479\u001b[0m \u001b[38;5;124;03mArgs:\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    486\u001b[0m \u001b[38;5;124;03m    direct feeds.\u001b[39;00m\n\u001b[0;32m    487\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m    488\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m graph\u001b[38;5;241m.\u001b[39mas_default():\n\u001b[1;32m--> 489\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fetch_mapper \u001b[38;5;241m=\u001b[39m \u001b[43m_FetchMapper\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfor_fetch\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfetches\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    490\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fetches \u001b[38;5;241m=\u001b[39m []\n\u001b[0;32m    491\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_targets \u001b[38;5;241m=\u001b[39m []\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:270\u001b[0m, in \u001b[0;36m_FetchMapper.for_fetch\u001b[1;34m(fetch)\u001b[0m\n\u001b[0;32m    266\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mArgument `fetch` = \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfetch\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m has invalid type \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m    267\u001b[0m                   \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mtype\u001b[39m(fetch)\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m. Cannot be None\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m    268\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(fetch, (\u001b[38;5;28mlist\u001b[39m, \u001b[38;5;28mtuple\u001b[39m)):\n\u001b[0;32m    269\u001b[0m   \u001b[38;5;66;03m# NOTE(touts): This is also the code path for namedtuples.\u001b[39;00m\n\u001b[1;32m--> 270\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_ListFetchMapper\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfetch\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    271\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(fetch, collections_abc\u001b[38;5;241m.\u001b[39mMapping):\n\u001b[0;32m    272\u001b[0m   \u001b[38;5;28;01mreturn\u001b[39;00m _DictFetchMapper(fetch)\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:382\u001b[0m, in \u001b[0;36m_ListFetchMapper.__init__\u001b[1;34m(self, fetches)\u001b[0m\n\u001b[0;32m    380\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m    381\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fetch_type \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mtype\u001b[39m(fetches)\n\u001b[1;32m--> 382\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mappers \u001b[38;5;241m=\u001b[39m \u001b[43m[\u001b[49m\u001b[43m_FetchMapper\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfor_fetch\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfetch\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mfor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mfetch\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mfetches\u001b[49m\u001b[43m]\u001b[49m\n\u001b[0;32m    383\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_unique_fetches, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_value_indices \u001b[38;5;241m=\u001b[39m _uniquify_fetches(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mappers)\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:382\u001b[0m, in \u001b[0;36m<listcomp>\u001b[1;34m(.0)\u001b[0m\n\u001b[0;32m    380\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m    381\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fetch_type \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mtype\u001b[39m(fetches)\n\u001b[1;32m--> 382\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mappers \u001b[38;5;241m=\u001b[39m [\u001b[43m_FetchMapper\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfor_fetch\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfetch\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;28;01mfor\u001b[39;00m fetch \u001b[38;5;129;01min\u001b[39;00m fetches]\n\u001b[0;32m    383\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_unique_fetches, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_value_indices \u001b[38;5;241m=\u001b[39m _uniquify_fetches(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_mappers)\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:280\u001b[0m, in \u001b[0;36m_FetchMapper.for_fetch\u001b[1;34m(fetch)\u001b[0m\n\u001b[0;32m    278\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(fetch, tensor_type):\n\u001b[0;32m    279\u001b[0m       fetches, contraction_fn \u001b[38;5;241m=\u001b[39m fetch_fn(fetch)\n\u001b[1;32m--> 280\u001b[0m       \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_ElementFetchMapper\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfetches\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcontraction_fn\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    281\u001b[0m \u001b[38;5;66;03m# Did not find anything.\u001b[39;00m\n\u001b[0;32m    282\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mArgument `fetch` = \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfetch\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m has invalid type \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m    283\u001b[0m                 \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mtype\u001b[39m(fetch)\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "File \u001b[1;32m~\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\tensorflow\\python\\client\\session.py:311\u001b[0m, in \u001b[0;36m_ElementFetchMapper.__init__\u001b[1;34m(self, fetches, contraction_fn)\u001b[0m\n\u001b[0;32m    308\u001b[0m   \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_unique_fetches\u001b[38;5;241m.\u001b[39mappend(ops\u001b[38;5;241m.\u001b[39mget_default_graph()\u001b[38;5;241m.\u001b[39mas_graph_element(\n\u001b[0;32m    309\u001b[0m       fetch, allow_tensor\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m, allow_operation\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m))\n\u001b[0;32m    310\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m--> 311\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mArgument `fetch` = \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfetch\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m has invalid type \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m    312\u001b[0m                   \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mtype\u001b[39m(fetch)\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m must be a string or Tensor. \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m    313\u001b[0m                   \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m(\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mstr\u001b[39m(e)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m)\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m    314\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[0;32m    315\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mArgument `fetch` = \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfetch\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m cannot be interpreted as \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m    316\u001b[0m                    \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124ma Tensor. (\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mstr\u001b[39m(e)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m)\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "\u001b[1;31mTypeError\u001b[0m: Argument `fetch` = <keras.src.optimizers.sgd.SGD object at 0x000001AD1B09E5D0> has invalid type \"SGD\" must be a string or Tensor. (Can not convert a SGD into a Tensor or Operation.)"
     ]
    }
   ],
   "source": [
    "seed = 42\n",
    "random.seed(seed)\n",
    "tf.random.set_seed(seed)\n",
    "np.random.seed(seed)\n",
    "\n",
    "# Measure Start Time\n",
    "start_time = time.time()\n",
    "\n",
    "### Training phase ###\n",
    "init = tf.compat.v1.global_variables_initializer()\n",
    "\n",
    "with tf.compat.v1.Session() as sess:\n",
    "    sess.run(init)\n",
    "    \n",
    "    # Collect list of generator/discriminator losses\n",
    "    gan_fraud_loss = []\n",
    "    \n",
    "    # Collect generated fraud data; Epoch as key, array as values\n",
    "    gan_fraud_X = {}\n",
    "    \n",
    "    ## Execute training epochs ##\n",
    "    for epoch in range(gan_epochs):\n",
    "        \n",
    "        # Feed X mini-batch\n",
    "        # Drop following variable 'Time'\n",
    "        split_df_dict = uniform_draw_feat_class(train_df, 'Class', mini_bat)\n",
    "        fraud_X_mb = split_df_dict[1].drop(columns=['V15', 'V13', 'V23', \n",
    "                                                    'Amount', 'V26', 'V22', 'Class']).to_numpy().astype(np.float32)\n",
    "    \n",
    "        # Feed prior(z)\n",
    "        fraud_z_sample = np.random.normal(size=[mini_bat, prior_z_dim]).astype(np.float32)\n",
    "        \n",
    "        # Minimise Losses\n",
    "        _, fraud_dloss = sess.run([gan_fraud.d_trainer_, gan_fraud.d_loss_], \n",
    "                                      feed_dict={gan_fraud.X_: fraud_X_mb, \n",
    "                                                 gan_fraud.prior_z_: fraud_z_sample})\n",
    "        _, fraud_gloss = sess.run([gan_fraud.g_trainer_, gan_fraud.g_loss_], \n",
    "                                      feed_dict={gan_fraud.X_: fraud_X_mb,\n",
    "                                                 gan_fraud.prior_z_: fraud_z_sample})\n",
    "        \n",
    "        # Store losses per epoch\n",
    "        gan_fraud_loss.append((fraud_dloss, fraud_gloss))\n",
    "        \n",
    "        if (epoch+1) % 100 == 0 or (epoch+1) == 1:\n",
    "            # Print generator, discriminator losses for fraudless & fraud cases\n",
    "            print('Epoch: {}'.format(epoch+1))\n",
    "            print('Fraud generator loss: {} | discriminator loss: {}'.format(fraud_gloss, fraud_dloss), '\\n')\n",
    "        \n",
    "        # Store generated data\n",
    "        fraud_z_gen = np.random.normal(size=[mini_bat, prior_z_dim]).astype(np.float32)\n",
    "        gan_fraud_X[epoch+1] = sess.run(gan_fraud.gen_X_, feed_dict={gan_fraud.prior_z_: fraud_z_gen})\n",
    "\n",
    "# Measure End Time\n",
    "elapsed_time = time.time() - start_time\n",
    "print('Time elapsed to train: ', elapsed_time)\n",
    "     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "498046a1-0eee-4be4-bfde-dacb20846e1e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "80a1f5d4-4805-427f-aba2-77b67b546874",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "55c195af-f42f-486f-9729-9b34a788f307",
   "metadata": {},
   "outputs": [],
   "source": [
    "# # Generate predictions on the test set\n",
    "# y_pred = discriminator.predict(X_test)\n",
    "\n",
    "# # Convert the predictions to binary (0 or 1)\n",
    "# y_pred_binary = (y_pred > 0.5).astype(int)\n",
    "\n",
    "# # Evaluate accuracy\n",
    "# accuracy = accuracy_score(y_test, y_pred_binary)\n",
    "# print(f\"Accuracy: {accuracy}\")\n",
    "\n",
    "# # Evaluate precision\n",
    "# precision = precision_score(y_test, y_pred_binary)\n",
    "# print(f\"Precision: {precision}\")\n",
    "\n",
    "# # Evaluate recall\n",
    "# recall = recall_score(y_test, y_pred_binary)\n",
    "# print(f\"Recall: {recall}\")\n",
    "\n",
    "# # Evaluate F1 score\n",
    "# f1 = f1_score(y_test, y_pred_binary)\n",
    "# print(f\"F1 Score: {f1}\")\n",
    "\n",
    "# # Display confusion matrix\n",
    "# conf_matrix = confusion_matrix(y_test, y_pred_binary)\n",
    "# print(\"Confusion Matrix:\")\n",
    "# print(conf_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ecf781a6-1f5a-428b-934d-64b50bc97388",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}