Neural Networks

Machine Learning for Finance in Python

Nathan George

Data Science Professor

GPU moore's law

Neural networks have potential

Neural nets have:

non-linearity
variable interactions
customizability

basic neural net diagram

net diagram with basic math

neural net math

activation neural net diagram

activation function

loss function

forward pass

backwards pass

keras and tensorflow logos

Implementing a neural net with keras

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense

Implementing a neural net with keras

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense


model = Sequential()

model.add(Dense(50,
                input_dim=scaled_train_features.shape[1],
                activation='relu'))
model.add(Dense(10, activation='relu'))
model.add(Dense(1, activation='linear'))

Fitting the model

model.compile(optimizer='adam', loss='mse')
history = model.fit(scaled_train_features, 
                    train_targets, 
                    epochs=50)

plt.plot(history.history['loss'])
plt.title('loss:' + str(round(history.history['loss'][-1], 6)))
plt.xlabel('epoch')
plt.ylabel('loss')
plt.show()

loss plot

Checking out performance

from sklearn.metrics import r2_score

# calculate R^2 score
train_preds = model.predict(scaled_train_features)
print(r2_score(train_targets, train_preds))

0.4771387560719418

Plot performance

# plot predictions vs actual
plt.scatter(train_preds, train_targets)
plt.xlabel('predictions')
plt.ylabel('actual')
plt.show()

preds vs actual

Make a neural net!

Machine Learning for Finance in Python