# a Hybrid Deep Learning Model

## Uses Supervised+Unsupervised Deep ML

# Part 1 - Identify the Frauds with the Self-Organizing Map

# Importing the libraries

import numpy as np

import matplotlib.pyplot as plt

import pandas as pd

# Importing the dataset

dataset = pd.read_csv('Credit_Card_Applications.csv')

X = dataset.iloc[:, :-1].values

y = dataset.iloc[:, -1].values

# Feature Scaling

from sklearn.preprocessing import MinMaxScaler

sc = MinMaxScaler(feature_range = (0, 1))

X = sc.fit_transform(X)

# Training the SOM

from minisom import MiniSom

som = MiniSom(x = 10, y = 10, input_len = 15, sigma = 1.0, learning_rate = 0.5)

som.random_weights_init(X)

som.train_random(data = X, num_iteration = 100)

# Visualizing the results

from pylab import bone, pcolor, colorbar, plot, show

bone()

pcolor(som.distance_map().T)

colorbar()

markers = ['o', 's']

colors = ['r', 'g']

for i, x in enumerate(X):

w = som.winner(x)

plot(w[0] + 0.5,

w[1] + 0.5,

markers[y[i]],

markeredgecolor = colors[y[i]],

markerfacecolor = 'None',

markersize = 10,

markeredgewidth = 2)

show()

# Finding the frauds

mappings = som.win_map(X)

frauds = np.concatenate((mappings[(5,5)], mappings[(1,1)]), axis = 0)

frauds = sc.inverse_transform(frauds)

# Part 2 - Going from Unsupervised to Supervised Deep Learning

# Creating the matrix of features

customers = dataset.iloc[:, 1:].values

# Creating the dependent variable--USing SOM output and dataset

is_fraud = np.zeros(len(dataset))

for i in range(len(dataset)):

if dataset.iloc[i,0] in frauds:#checks common..#using SOM result of customers we take common between som result and Dataset

is_fraud[i] = 1

# Feature Scaling

from sklearn.preprocessing import StandardScaler

sc = StandardScaler()

customers = sc.fit_transform(customers)

# Part 2 - Making a ANN!

# Importing the Keras libraries and packages

from keras.models import Sequential

from keras.layers import Dense

# Initialising the ANN

classifier = Sequential()

# Adding the input layer and the first hidden layer

classifier.add(Dense(units = 8 , kernel_initializer = 'uniform', activation = 'relu', input_dim = 15))

# Adding another hidden layer

classifier.add(Dense(units = 8 , kernel_initializer = 'uniform', activation = 'relu'))

# Adding the output layer

classifier.add(Dense(units = 1, kernel_initializer = 'uniform', activation = 'sigmoid'))

# Compiling the ANN

classifier.compile(optimizer = 'adam', loss = 'binary_crossentropy', metrics = ['accuracy'])

# Fitting the ANN to the Training set

classifier.fit(customers, is_fraud, batch_size = 10, epochs = 20)

# Predicting the probabilities of frauds

y_pred = classifier.predict(customers)

y_pred = np.concatenate((dataset.iloc[:, 0:1].values, y_pred), axis = 1) # row wise concatenate of customer id and probablity of fraus

y_pred = y_pred[y_pred[:, 1].argsort()]