def neural_network_regression(data):
layers0 = [('input', InputLayer),
('dense0', DenseLayer),
('dense1', DenseLayer),
('dense2', DenseLayer),
('dense3', DenseLayer),
('dense4', DenseLayer),
('output', DenseLayer)]
net0 = NeuralNet(layers=layers0,
input_shape=(None, len(data.X_train[0])),
dense0_num_units=400,
dense0_nonlinearity=rectify,
dense1_num_units=200,
dense1_nonlinearity=rectify,
dense2_num_units=100,
dense2_nonlinearity=rectify,
dense3_num_units=50,
dense3_nonlinearity=rectify,
dense4_num_units=25,
dense4_nonlinearity=rectify,
output_num_units=1,
output_nonlinearity=None,
update=nesterov_momentum,
update_learning_rate=0.00001,
update_momentum=0.9,
regression=True,
on_epoch_finished=[
EarlyStopping(patience=20),
AcceptLoss(min=0.01)
],
verbose=VERBOSE,
max_epochs=100000)
# Provide our own validation set
def my_split(self, X, y, eval_size):
return data.X_train, data.X_validate, data.y_train_nn, data.y_validate_nn
net0.train_split = types.MethodType(my_split, net0)
return net0
# Compute the z-scores for both train and validation. However, use mean and standard deviation for training
# on both. This is customary because we trained on this standard deviation and mean. Additionally, our
# prediction set might too small to calculate a meaningful mean and standard deviation.
X_train_z = zscore(X_train, train_mean, train_sdev) #scipy.stats.mstats.zscore(X_train)
X_validate_z = zscore(X_validate, train_mean, train_sdev) #scipy.stats.mstats.zscore(X_validate)
#These can be used to check my zscore calc to numpy
#print(X_train_z)
#print(scipy.stats.mstats.zscore(X_train))
# Provide our own validation set
def my_split(self, X, y, eval_size):
return X_train_z,X_validate_z,y_train,y_validate
net0.train_split = types.MethodType(my_split, net0)
# Train the network
net0.fit(X_train_z,y_train)
# Predict the validation set
pred_y = net0.predict(X_validate_z)
# Display predictions and count the number of incorrect predictions.
species_names = ['setosa','versicolour','virginica']
count = 0
wrong = 0
for element in zip(X_validate,y_validate,pred_y):
print("Input: sepal length: {}, sepal width: {}, petal length: {}, petal width: {}; Expected: {}; Actual: {}".format(
element[0][0],element[0][1],element[0][2],element[0][3],
