def train_model(self, X, y, batch_size=1, epochs=10):
"""Train a Convolutional Neural Network to recognize handwritten characters.
Args:
X (numpy.ndarray)8: Training data (EMNIST ByClass dataset)
y (numpy.ndarray): Labels of the training data.
batch_size (int): How many images the CNN should use at a time.
epochs (int): How many times the data should be used to train the model.
"""
self.build_model()
callbacks.Callback()
checkpoint = callbacks.ModelCheckpoint(filepath=getcwd() +
'weights.{epoch:02d}.hdf5',
monitor='val_loss',
verbose=0,
save_best_only=False,
save_weights_only=False,
mode='auto',
period=1)
#self.model.load_weights(filepath=getcwd()+'weights.04.hdf5', by_name=False)
callbacks.TensorBoard(log_dir=getcwd() + 'weights.{epoch:02d}')
self.model.fit(X,
y,
batch_size=batch_size,
epochs=epochs,
callbacks=[checkpoint])
def train_model(model, epochs, x_test, y_test, x_train, y_train, weights_path,
log_path, learning_rate):
"""
Training a model on training data, while evaluating it on the validation set
Args:
model: Keras model to train
epochs: number of epochs the model should be trained for
x_train: training data
x_test: validation data
y_train: training labels
y_test: validation labels
weights_path: the filepath of weights of the model, if existant
log_path: filepath of were to log data about training into
learning_rate: learning rate used to train the model
Returns:
-
"""
adam = optimizers.Adam(lr=learning_rate)
sdg = optimizers.SGD(lr=learning_rate,
momentum=0.0,
decay=0.0,
nesterov=False)
model.compile(loss='categorical_crossentropy',
optimizer=adam,
metrics=['accuracy'])
if os.path.isfile(weights_path):
model.load_weights(weights_path, by_name=False)
cb = callbacks.Callback()
cb.set_model(model)
tensorboard = TensorBoard(log_dir=log_path)
tensorboard.set_model(model)
checkpoint = ModelCheckpoint(weights_path,
monitor='val_acc',
verbose=1,
save_best_only=False,
mode='max')
#early_stop = EarlyStopping(monitor='val_loss', patience=5, verbose=0)
callbacks_list = [checkpoint, cb, tensorboard]
model.fit(x_train,
y_train,
batch_size=256,
epochs=epochs,
shuffle=True,
verbose=2,
callbacks=callbacks_list,
validation_data=(x_test, y_test))
model.save_weights(weights_path)
score = model.evaluate(x_test, y_test, batch_size=128)
print(score)
def train(self, data, labels, batch_size=128, epochs=2):
#Il modello viene allenato
self.getModel()
callbacks.Callback()
checkpoint = callbacks.ModelCheckpoint(filepath=getcwd() +
'weights.{epoch:02d}.hdf5',
monitor='val_loss',
verbose=0,
save_best_only=False,
save_weights_only=False,
mode='auto',
period=1)
# self.model.load_weights(filepath=getcwd()+'weights.10.hdf5', by_name=False)
self.model.fit(data,
labels,
batch_size=batch_size,
epochs=epochs,
callbacks=[checkpoint])
print(checkpoint)
# print(Counter(X_train))
print("---" * 80)
c = 1
for an in Counter(ans):
if c <= 6:
print(an)
c += 1
c = 1
for an in Counter(SP):
if c <= 1:
print("SP:", an)
c += 1
# print(len(SP))
# print(len(ans))
# print(len(X_train))
callbacks.Callback()
callbacks.History()
callbacks.TensorBoard(log_dir='./logs',
histogram_freq=0,
batch_size=32,
write_graph=True,
write_grads=False,
write_images=False,
embeddings_freq=0,
embeddings_layer_names=None,
embeddings_metadata=None,
embeddings_data=None,
update_freq='epoch')
callbacks.ReduceLROnPlateau(monitor='val_loss',
factor=0.1,
patience=10,