0.2, # randomly shift images horizontally (fraction of total width)
height_shift_range=
0.2, # randomly shift images vertically (fraction of total height)
horizontal_flip=True, # randomly flip images
vertical_flip=False) # randomly flip images
datagen.fit(x_train)
#train
batch_size = 16
epochs = 30
if (options.validation):
epochs = 50
History = model.fit_generator(datagen.flow(x_train,
y_train,
batch_size=batch_size),
epochs=epochs,
validation_data=(x_test, y_test),
verbose=1,
steps_per_epoch=x_train.shape[0] //
batch_size)
else:
History = model.fit_generator(datagen.flow(x_train,
y_train,
batch_size=batch_size),
epochs=epochs,
verbose=1,
steps_per_epoch=x_train.shape[0] //
batch_size)
#generate submission
if (options.save_submission):
if remote_execution:
class LogRunMetrics(Callback):
# callback at the end of every epoch
def on_epoch_end(self, epoch, log):
# log a value repeated which creates a list
run.log('val_loss', log['val_loss'])
run.log('loss', log['loss'])
callbacks.append(LogRunMetrics())
model.fit_generator(
train_generator,
steps_per_epoch=5,
epochs=3,
verbose=1,
# steps_per_epoch=50, epochs=30, verbose=1,
callbacks=callbacks,
validation_data=val_generator,
validation_steps=80,
workers=4)
# # Loss/epoch plots
plt.plot(model.history.history['categorical_crossentropy'], label='train')
plt.plot(model.history.history['val_categorical_crossentropy'], label='val')
plt.legend()
plt.xlabel('epoch')
plt.ylabel('logloss')
# log this plot to the aml workspace so we can see it in the azure portal
if remote_execution:
run.log_image('logloss', plot=plt)
