def train():
# treina a NN usando os dados gerados
nn_input, expected_output = load_data('train')
output_len = len(expected_output[0])
height = len(nn_input[0])
width = len(nn_input[0][0])
model = inverse_planning_model(width=width,
height=height,
output_len=output_len)
model.compile(loss=keras.losses.categorical_crossentropy,
optimizer=keras.optimizers.Adam(),
metrics=['accuracy'])
model.summary()
tensorboard = TensorBoard(log_dir=os.path.join("logs",
"{}".format(time())),
profile_batch=0)
model.fit(nn_input,
expected_output,
batch_size=batch_size,
epochs=num_epochs,
callbacks=[tensorboard])
save_model_to_json(model, 'inverse_planning_model')
def train(model, network_input, network_output, X_train, X_test, y_train, y_test, results_dir):
callbacks_list = utils.model_callbacks(results_dir)
utils.logging('Loaded model callbacks')
utils.save_model_to_json(model, results_dir)
utils.logging('Model saved to file: {}/{}'.format(results_dir, 'model.json'))
history = model.fit(network_input, network_output,
validation_data=(X_test, y_test),
validation_split=0.33,
epochs=200,
batch_size=64,
callbacks=callbacks_list,
verbose=1,
)
utils.generate_final_plots(history, results_dir)
BATCH_SIZE = 128
train_features, train_labels = read_mnist('train-images-idx3-ubyte.gz', 'train-labels-idx1-ubyte.gz')
train_features, validation_features, train_labels, validation_labels = \
train_test_split(train_features, train_labels, test_size=0.2, random_state=0)
print('# of training images:', train_features.shape[0])
print('# of cross-validation images:', validation_features.shape[0])
model = make_lenet5()
model.summary()
model.compile(loss=keras.losses.categorical_crossentropy, optimizer=keras.optimizers.Adam(), metrics=['accuracy'])
X_train, y_train = train_features, to_categorical(train_labels)
X_validation, y_validation = validation_features, to_categorical(validation_labels)
train_generator = ImageDataGenerator().flow(X_train, y_train, batch_size=BATCH_SIZE)
validation_generator = ImageDataGenerator().flow(X_validation, y_validation, batch_size=BATCH_SIZE)
steps_per_epoch = X_train.shape[0] // BATCH_SIZE
validation_steps = X_validation.shape[0] // BATCH_SIZE
tensorboard = TensorBoard(log_dir=os.path.join("logs", "{}".format(time())))
model.fit(train_generator, steps_per_epoch=steps_per_epoch, epochs=EPOCHS,
validation_data=validation_generator, validation_steps=validation_steps,
shuffle=True, callbacks=[tensorboard])
save_model_to_json(model, 'lenet5')
print("before CNN")
model = CNN_Model(num_samples, input_volume=3).get_model()
print("after CNN")
# es_callback = EarlyStopping(monitor='val_loss',patience=10,restore_best_weights=True)
history = model.fit(
input_train,
output_train,
epochs=num_epochs,
verbose=1,
validation_data=(input_validation,
output_validation)) #, callbacks=[es_callback])
save_model_to_json(model, json_file_name)
# summarize history for loss
plt.rcParams.update({'font.size': 14})
plt.figure()
plt.grid()
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
# plt.title('model loss')
plt.ylabel('Loss')
plt.xlabel('Epoch')
plt.legend(['Train', 'Validation'], loc='upper left')
plt.savefig('train_val.eps', format='eps')
plt.show()