def test_random_translation_negative_lower(self):
mock_offset = np.random.random((12, 1))
with test.mock.patch.object(
gen_stateful_random_ops, 'stateful_uniform', return_value=mock_offset):
with self.cached_session(use_gpu=True):
layer = image_preprocessing.RandomTranslation((-0.2, .3), .4)
layer_2 = image_preprocessing.RandomTranslation((0.2, .3), .4)
inp = np.random.random((12, 5, 8, 3)).astype(np.float32)
actual_output = layer(inp, training=1)
actual_output_2 = layer_2(inp, training=1)
self.assertAllClose(actual_output, actual_output_2)
def test_config_with_custom_name(self):
layer = image_preprocessing.RandomTranslation(.5,
.6,
name='image_preproc')
config = layer.get_config()
layer_1 = image_preprocessing.RandomTranslation.from_config(config)
self.assertEqual(layer_1.name, layer.name)
def test_distribution(self, distribution):
if "CentralStorage" in type(distribution).__name__:
self.skipTest("Does not work with CentralStorageStrategy yet.")
# TODO(b/159738418): large image input causes OOM in ubuntu multi gpu.
np_images = np.random.random((32, 32, 32, 3)).astype(np.float32)
image_dataset = dataset_ops.Dataset.from_tensor_slices(
np_images).batch(16, drop_remainder=True)
with distribution.scope():
input_data = keras.Input(shape=(32, 32, 3), dtype=dtypes.float32)
image_preprocessor = keras.Sequential([
image_preprocessing.Resizing(height=256, width=256),
image_preprocessing.RandomCrop(height=224, width=224),
image_preprocessing.RandomTranslation(.1, .1),
image_preprocessing.RandomRotation(.2),
image_preprocessing.RandomFlip(),
image_preprocessing.RandomZoom(.2, .2)
])
preprocessed_image = image_preprocessor(input_data)
flatten_layer = keras.layers.Flatten(data_format="channels_last")
output = flatten_layer(preprocessed_image)
cls_layer = keras.layers.Dense(units=1, activation="sigmoid")
output = cls_layer(output)
model = keras.Model(inputs=input_data, outputs=output)
model.compile(loss="binary_crossentropy")
_ = model.predict(image_dataset)
def test_random_translation_inference(self):
with CustomObjectScope(
{'RandomTranslation': image_preprocessing.RandomTranslation}):
input_images = np.random.random((2, 5, 8, 3)).astype(np.float32)
expected_output = input_images
with tf_test_util.use_gpu():
layer = image_preprocessing.RandomTranslation(.5, .5)
actual_output = layer(input_images, training=0)
self.assertAllClose(expected_output, actual_output)
def test_random_translation_left_numeric_constant(self):
for dtype in (np.int64, np.float32):
with tf_test_util.use_gpu():
input_image = np.reshape(np.arange(0, 25), (1, 5, 5, 1)).astype(dtype)
# Shifting by -.2 * 5 = 1 pixel.
layer = image_preprocessing.RandomTranslation(
height_factor=0., width_factor=(-.2, -.2), fill_mode='constant')
output_image = layer(input_image)
# pyformat: disable
expected_output = np.asarray([
[1, 2, 3, 4, 0],
[6, 7, 8, 9, 0],
[11, 12, 13, 14, 0],
[16, 17, 18, 19, 0],
[21, 22, 23, 24, 0]
]).astype(dtype)
# pyformat: enable
expected_output = np.reshape(expected_output, (1, 5, 5, 1))
self.assertAllEqual(expected_output, output_image)
def test_random_translation_down_numeric_reflect(self):
for dtype in (np.int64, np.float32):
with tf_test_util.use_gpu():
input_image = np.reshape(np.arange(0, 25), (1, 5, 5, 1)).astype(dtype)
# Shifting by .2 * 5 = 1 pixel.
layer = image_preprocessing.RandomTranslation(
height_factor=(.2, .2), width_factor=0.)
output_image = layer(input_image)
# pyformat: disable
expected_output = np.asarray([
[0, 1, 2, 3, 4],
[0, 1, 2, 3, 4],
[5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]
]).astype(dtype)
# pyformat: enable
expected_output = np.reshape(expected_output, (1, 5, 5, 1))
self.assertAllEqual(expected_output, output_image)
def test_random_translation_asymmetric_size_numeric_reflect(self):
for dtype in (np.int64, np.float32):
with tf_test_util.use_gpu():
input_image = np.reshape(np.arange(0, 16), (1, 8, 2, 1)).astype(dtype)
# Shifting by .5 * 8 = 1 pixel.
layer = image_preprocessing.RandomTranslation(
height_factor=(.5, .5), width_factor=0.)
output_image = layer(input_image)
# pyformat: disable
expected_output = np.asarray([
[6, 7],
[4, 5],
[2, 3],
[0, 1],
[0, 1],
[2, 3],
[4, 5],
[6, 7],
]).astype(dtype)
# pyformat: enable
expected_output = np.reshape(expected_output, (1, 8, 2, 1))
self.assertAllEqual(expected_output, output_image)
def test_distribution(self, distribution):
np_images = np.random.random((1000, 32, 32, 3)).astype(np.float32)
image_dataset = dataset_ops.Dataset.from_tensor_slices(
np_images).batch(32, drop_remainder=True)
with distribution.scope():
input_data = keras.Input(shape=(32, 32, 3), dtype=dtypes.float32)
image_preprocessor = keras.Sequential([
image_preprocessing.Resizing(height=256, width=256),
image_preprocessing.RandomCrop(height=224, width=224),
image_preprocessing.RandomTranslation(.1, .1),
image_preprocessing.RandomRotation(.2),
image_preprocessing.RandomFlip(),
image_preprocessing.RandomZoom(.2, .2)
])
preprocessed_image = image_preprocessor(input_data)
flatten_layer = keras.layers.Flatten(data_format="channels_last")
output = flatten_layer(preprocessed_image)
cls_layer = keras.layers.Dense(units=1, activation="sigmoid")
output = cls_layer(output)
model = keras.Model(inputs=input_data, outputs=preprocessed_image)
model.compile(loss="binary_crossentropy")
_ = model.predict(image_dataset)
from multiprocessing.pool import ThreadPool
from tqdm import tqdm
import tensorflow as tf # 2.4
from tensorflow.keras import layers
from tensorflow.keras.models import Sequential
from tensorflow.python.keras.layers.preprocessing import image_preprocessing
from tensorflow.keras.applications import EfficientNetB1, Xception
# 数据增强
image_augmentation = Sequential(
[
image_preprocessing.RandomRotation(factor=0.1), # 随机旋转
image_preprocessing.RandomTranslation(height_factor=0.1,
width_factor=0.1), # 随机平移
image_preprocessing.RandomFlip(), # 随机翻转
image_preprocessing.RandomContrast(factor=0.1), # 随机改变对比度
image_preprocessing.RandomZoom(height_factor=0.1,
width_factor=0.1), # 随机缩放
# image_preprocessing.RandomHeight(factor=0.1), # 随机改变高度
# image_preprocessing.RandomWidth(factor=0.1), # 随机改变宽度
# image_preprocessing.RandomCrop(height, width), # 随机裁剪
# image_preprocessing.CenterCrop(height, width), # 中心裁剪
],
name="img_augmentation",
)
def get_time_suffix(suffix_len=15):
"""获取一个时间后缀"""