test_labels_p, train_labels_p = np.split(labels_process_r, [test_index])
# Normalizing sets
if False:
train_data = scaler.fit_transform(train_data)
test_data = scaler.transform(test_data)
# Traning & Testing Model with Full labels
model = new_model(nb_classes)
model.fit(x=train_data,
y=train_labels_f,
batch_size=64,
epochs=5,
shuffle=True,
verbose=0)
acc_full.append(model.evaluate(x=test_data, y=test_labels_f, verbose=0)[1])
# Traning & Testing Model with Merged labels
model = new_model(int(nb_classes / 2))
model.fit(x=train_data,
y=train_labels_m,
batch_size=64,
epochs=5,
shuffle=True,
verbose=0)
acc_merged.append(
model.evaluate(x=test_data, y=test_labels_m, verbose=0)[1])
# Training & Testing Model with Process labels
model = new_model(3)
model.fit(x=train_data,
