In [81]:

    

import numpy as np
from data_prep import features, targets, features_test, targets_test


    

In [82]:

    

#print features


    

In [83]:

    

#print targets


    

In [84]:

    

#print features_test


    

In [85]:

    

#print targets_test


    

In [86]:

    

# Defining the sigmoid function for activations
def sigmoid(x):
    return 1 / (1 + np.exp(-x))


    

In [87]:

    

# Use to same seed to make debugging easier
np.random.seed(42)


    

In [88]:

    

#print features.shape
n_records, n_features = features.shape
#print n_records, n_features
last_loss = None


    

In [89]:

    

# Initialize weights
weights = np.random.normal(scale=1 / n_features**.5, size=n_features)
#print weights


    

In [90]:

    

# Neural Network hyperparameters
epochs = 5000
learnrate = 0.05


    

In [91]:

    

for e in range(epochs):
    del_w = np.zeros(weights.shape)
    for x, y in zip(features.values, targets):
        # Loop through all records, x is the input, y is the target

        # Note: We haven't included the h variable from the previous
        #       lesson. You can add it if you want, or you can calculate
        #       the h together with the output
        
        h = np.dot(x,weights)

        # TODO: Calculate the output
        output = sigmoid(h)

        # TODO: Calculate the error
        error = y - output

        # TODO: Calculate the error term
        error_term = error * output * (1 - output)

        # TODO: Calculate the change in weights for this sample
        #       and add it to the total weight change
        del_w += error_term * x

    # TODO: Update weights using the learning rate and the average change in weights
    weights += learnrate * del_w / n_records

    # Printing out the mean square error on the training set
    if e % (epochs / 10) == 0:
        out = sigmoid(np.dot(features, weights))
        loss = np.mean((out - targets) ** 2)
        if last_loss and last_loss < loss:
            print("Train loss: ", loss, "  WARNING - Loss Increasing")
        else:
            print("Train loss: ", loss)
        last_loss = loss


    

    




('Train loss: ', 0.2640432346990609)
('Train loss: ', 0.22373253255332445)
('Train loss: ', 0.2095067978359536)
('Train loss: ', 0.20364221702453061)
('Train loss: ', 0.20090206502342442)
('Train loss: ', 0.199466585743907)
('Train loss: ', 0.1986432137900257)
('Train loss: ', 0.1981368460789586)
('Train loss: ', 0.19780830304192196)
('Train loss: ', 0.19758610347010597)

In [92]:

    

# Calculate accuracy on test data
tes_out = sigmoid(np.dot(features_test, weights))
predictions = tes_out > 0.5
accuracy = np.mean(predictions == targets_test)
print("Prediction accuracy: {:.3f}".format(accuracy))


    

    




Prediction accuracy: 0.725

In [ ]: