def test_movie_data_generator_fit(self):
generator = image_generators.MovieDataGenerator(
featurewise_center=True,
samplewise_center=True,
featurewise_std_normalization=True,
samplewise_std_normalization=True,
zca_whitening=True,
data_format='channels_last')
# Test grayscale
x = np.random.random((8, 5, 10, 10, 1))
generator.fit(x)
# Test RBG
x = np.random.random((8, 5, 10, 10, 3))
generator.fit(x)
generator = image_generators.MovieDataGenerator(
featurewise_center=True,
samplewise_center=True,
featurewise_std_normalization=True,
samplewise_std_normalization=True,
zca_whitening=False,
data_format='channels_first')
# Test grayscale
x = np.random.random((8, 1, 5, 4, 6))
generator.fit(x)
# Test RBG
x = np.random.random((8, 3, 5, 4, 6))
generator.fit(x)
def test_movie_data_generator_invalid_data(self):
generator = image_generators.MovieDataGenerator(
featurewise_center=True,
samplewise_center=True,
featurewise_std_normalization=True,
samplewise_std_normalization=True,
zca_whitening=True,
data_format='channels_last')
# Test fit with invalid data
with self.assertRaises(ValueError):
x = np.random.random((3, 10, 10))
generator.fit(x)
# Test flow with invalid dimensions
with self.assertRaises(ValueError):
train_dict = {
'X': np.random.random((32, 10, 10, 1)),
'y': np.random.random((32, 10, 10, 1))
}
generator.flow(train_dict)
# Test flow with non-matching batches
with self.assertRaises(Exception):
train_dict = {
'X': np.random.random((32, 30, 10, 10, 1)),
'y': np.random.random((25, 30, 10, 10, 1))
}
generator.flow(train_dict)
# Test flow with bigger frames_per_batch than frames
with self.assertRaises(Exception):
train_dict = {
'X': np.random.random((32, 30, 10, 10, 1)),
'y': np.random.random((32, 30, 10, 10, 1))
}
generator.flow(train_dict, frames_per_batch=31)
# Invalid number of channels: will work but raise a warning
x = np.random.random((32, 30, 10, 10, 5))
y = np.random.random((32, 30, 10, 10, 1))
generator.flow({'X': x, 'y': y})
with self.assertRaises(ValueError):
generator = image_generators.MovieDataGenerator(
data_format='unknown')
generator = image_generators.MovieDataGenerator(zoom_range=(2, 2))
with self.assertRaises(ValueError):
generator = image_generators.MovieDataGenerator(zoom_range=(2, 2,
2))
def test_movie_data_generator(self):
frames = 7
for test_images in _generate_test_images():
img_list = []
for im in test_images:
frame_list = []
for _ in range(frames):
frame_list.append(img_to_array(im)[None, ...])
img_stack = np.vstack(frame_list)
img_list.append(img_stack)
images = np.vstack(img_list)
batches = images.shape[0] // frames
images = np.reshape(images, (batches, frames, *images.shape[1:]))
generator = image_generators.MovieDataGenerator(
featurewise_center=True,
samplewise_center=True,
featurewise_std_normalization=True,
samplewise_std_normalization=True,
zca_whitening=False,
rotation_range=90.,
width_shift_range=0.1,
height_shift_range=0.1,
shear_range=0.5,
zoom_range=0.2,
channel_shift_range=1.,
brightness_range=(1, 5),
fill_mode='nearest',
cval=0.5,
horizontal_flip=True,
vertical_flip=True)
# Basic test before fit
train_dict = {
'X': np.random.random((8, 11, 10, 10, 3)),
'y': np.random.random((8, 11, 10, 10, 1)),
}
generator.flow(train_dict, frames_per_batch=1)
# Temp dir to save generated images
temp_dir = self.get_temp_dir()
# Fit
generator.fit(images, augment=True, seed=1)
y_shape = tuple(list(images.shape)[:-1] + [1])
train_dict['X'] = images
train_dict['y'] = np.random.random(y_shape)
frames_per_batch = 5
for x, y in generator.flow(train_dict,
shuffle=True,
save_to_dir=temp_dir,
skip=1,
frames_per_batch=frames_per_batch):
batch_x_shape = (frames_per_batch, *images.shape[2:])
batch_y_shape = (frames_per_batch, *y_shape[2:])
self.assertEqual(x.shape[1:], batch_x_shape)
self.assertEqual(len(y), 2)
self.assertEqual(y[0].shape[1:], batch_y_shape)
self.assertEqual(y[-1].shape[1:], batch_y_shape)
break
def test_batch_standardize(self):
# MovieDataGenerator.standardize should work on batches
frames = 3
for test_images in _generate_test_images():
img_list = []
for im in test_images:
frame_list = []
for _ in range(frames):
frame_list.append(img_to_array(im)[None, ...])
img_stack = np.vstack(frame_list)
img_list.append(img_stack)
images = np.vstack(img_list)
batches = images.shape[0] // frames
images = np.reshape(images, (batches, frames, *images.shape[1:]))
generator = image_generators.MovieDataGenerator(
featurewise_center=True,
samplewise_center=True,
featurewise_std_normalization=True,
samplewise_std_normalization=True,
zca_whitening=False,
rescale=2,
preprocessing_function=lambda x: x,
rotation_range=90.,
width_shift_range=0.1,
height_shift_range=0.1,
shear_range=0.5,
zoom_range=0.2,
channel_shift_range=0.,
brightness_range=(1, 5),
fill_mode='nearest',
cval=0.5,
horizontal_flip=True,
vertical_flip=True)
generator.fit(images, augment=True)
transformed = np.copy(images)
for i, im in enumerate(transformed):
transformed[i] = generator.random_transform(im, seed=1)
transformed = generator.standardize(transformed)
