print("Output class: ", np.argmax(workspace.FetchBlob("softmax"), axis=1))
print("Real class : ", label)
if __name__ == '__main__':
# 1. Set global init level & Device Option: CUDA or CPU
core.GlobalInit(['caffe2', '--caffe2_log_level=0'])
if args.use_gpu:
device_opts = core.DeviceOption(caffe2_pb2.CUDA, 0)
else:
device_opts = core.DeviceOption(caffe2_pb2.CPU, 0)
# 2. Prepare data
# try to download & extract
# then do shuffle & -std/mean normalization
train_x, train_y, test_x, test_y = prepare_data()
train_x, test_x = normalization(train_x, test_x)
# 3. Start training & save pb files.
do_train(train_x,
train_y,
test_x,
test_y,
epochs=args.epochs,
device_opts=device_opts,
use_gpu=args.use_gpu)
# 4. Do a test if you need
do_test(test_x, test_y, device_opts)
def train(model):
batch_size = 64
num_classes = 10
epochs = 100
x_train, y_train, x_test, y_test = data_utility.prepare_data()
# x_train = x_train.astype('float32')
# x_test = x_test.astype('float32')
x_train, x_test = x_train / 255.0, x_test / 255.0
print('x_train shape:', x_train.shape, 'x_test shape:', x_test.shape)
# y_train = utils.to_categorical(y_train, num_classes)
# y_test = utils.to_categorical(y_test, num_classes)
# sgd = SGD(lr=0.0001, decay=1e-8, momentum=0.9, nesterov=True)
model.compile(optimizer='adam',
loss='categorical_crossentropy',
metrics=['accuracy'])
model.summary()
data_augmentation = True
if not data_augmentation:
print('Not using data augmentation.')
model.fit(x_train,
y_train,
batch_size=batch_size,
epochs=epochs,
validation_data=(x_test, y_test),
shuffle=True)
else:
print('Using real-time data augmentation.')
# This will do preprocessing and realtime data augmentation:
datagen = ImageDataGenerator(
featurewise_center=False, # set input mean to 0 over the dataset
samplewise_center=False, # set each sample mean to 0
featurewise_std_normalization=False, # divide inputs by dataset std
samplewise_std_normalization=False, # divide each input by its std
zca_whitening=False, # apply ZCA whitening
zca_epsilon=1e-06, # epsilon for ZCA whitening
rotation_range=0, # randomly rotate images in 0 to 180 degrees
width_shift_range=0.1, # randomly shift images horizontally
height_shift_range=0.1, # randomly shift images vertically
shear_range=0., # set range for random shear
zoom_range=0., # set range for random zoom
channel_shift_range=0., # set range for random channel shifts
# set mode for filling points outside the input boundaries
fill_mode='nearest',
cval=0., # value used for fill_mode = "constant"
horizontal_flip=True, # randomly flip images
vertical_flip=False, # randomly flip images
# set rescaling factor (applied before any other transformation)
rescale=None,
# set function that will be applied on each input
preprocessing_function=None,
# image data format, either "channels_first" or "channels_last"
data_format=None,
# fraction of images reserved for validation (strictly between 0 and 1)
validation_split=0.0)
# Compute quantities required for feature-wise normalization
# (std, mean, and principal components if ZCA whitening is applied).
datagen.fit(x_train)
# Fit the model on the batches generated by datagen.flow().
model.fit_generator(datagen.flow(x_train,
y_train,
batch_size=batch_size),
epochs=epochs,
validation_data=(x_test, y_test),
workers=4)
label_path)
continue
try:
tf_example = dict_to_tf_example(image_path, label_path)
writer.write(tf_example.SerializeToString())
except ValueError:
tf.logging.warning('Invalid example: %s, ignoring.', example)
writer.close()
if __name__ == "__main__":
# debug
seq_list = load_data_names("/cvgl/group/GazeCapture/test")
batch_size = len(seq_list)
dataset_path = "/media/insfan/00028D8D000E9194/MPIIFaceGaze/MPIIFaceGaze/MPIIFaceGaze_fem64.npz"
train_data = load_data(dataset_path)
data_utility.prepare_data(train_data)
# img_ch = 3
# img_cols = 64
# img_rows = 64
#
# test_batch = load_batch_from_names_random(seq_list, dataset_path, batch_size, 64, 64, 3)
#
# print("Loaded: {} data".format(len(test_batch[0][0])))