def _augment(self, img, _):
h, w = img.shape[:2]
area = h * w
for _ in range(10):
targetArea = self.rng.uniform(self.crop_area_fraction, 1.0) * area
aspectR = self.rng.uniform(self.aspect_ratio_low,
self.aspect_ratio_high)
ww = int(np.sqrt(targetArea * aspectR) + 0.5)
hh = int(np.sqrt(targetArea / aspectR) + 0.5)
if self.rng.uniform() < 0.5:
ww, hh = hh, ww
if hh <= h and ww <= w:
x1 = 0 if w == ww else self.rng.randint(0, w - ww)
y1 = 0 if h == hh else self.rng.randint(0, h - hh)
out = img[y1:y1 + hh, x1:x1 + ww]
out = cv2.resize(out, (self.target_shape, self.target_shape),
interpolation=cv2.INTER_CUBIC)
return out
out = imgaug.ResizeShortestEdge(self.target_shape,
interp=cv2.INTER_CUBIC).augment(img)
out = imgaug.CenterCrop(self.target_shape).augment(out)
return out
def get_data(datadir, size=IMAGESIZE, isTrain=True, zmin=-1, zmax=1):
if isTrain:
augs = [
imgaug.ResizeShortestEdge(int(size * 1.143)),
imgaug.RandomCrop(size),
imgaug.Flip(horiz=True),
]
else:
augs = [
imgaug.ResizeShortestEdge(int(size * 1.143)),
imgaug.CenterCrop(size)
]
def get_images(dir):
files = sorted(glob.glob(os.path.join(dir, "*.jpg")))
df = ImageFromFile(files, channel=3, shuffle=isTrain)
random_df = RandomZData([size, size, 3], zmin, zmax)
return JoinData([random_df, AugmentImageComponent(df, augs)])
names = ['train'] if isTrain else ['test']
df = get_images(*[os.path.join(datadir, n) for n in names])
df = BatchData(df, BATCH if isTrain else TEST_BATCH)
return df
def fbresnet_augmentor(isTrain):
"""
Augmentor used in fb.resnet.torch, for BGR images in range [0,255].
"""
if isTrain:
"""
Sec 5.1:
We use scale and aspect ratio data augmentation [35] as
in [12]. The network input image is a 224×224 pixel random
crop from an augmented image or its horizontal flip.
"""
augmentors = [
imgaug.GoogleNetRandomCropAndResize(interp=cv2.INTER_LINEAR),
# It's OK to remove the following augs if your CPU is not fast enough.
# Removing brightness/contrast/saturation does not have a significant effect on accuracy.
# Removing lighting leads to a tiny drop in accuracy.
imgaug.RandomOrderAug([
imgaug.BrightnessScale((0.6, 1.4), clip=False),
imgaug.Contrast((0.6, 1.4), rgb=False, clip=False),
imgaug.Saturation(0.4, rgb=False),
# rgb-bgr conversion for the constants copied from fb.resnet.torch
imgaug.Lighting(
0.1,
eigval=np.asarray([0.2175, 0.0188, 0.0045][::-1]) * 255.0,
eigvec=np.array([[-0.5675, 0.7192, 0.4009],
[-0.5808, -0.0045, -0.8140],
[-0.5836, -0.6948, 0.4203]],
dtype='float32')[::-1, ::-1])
]),
imgaug.Flip(horiz=True),
]
else:
augmentors = [
imgaug.ResizeShortestEdge(256, cv2.INTER_LINEAR),
imgaug.CenterCrop((224, 224)),
]
return augmentors
def get_data(datadir,
size=IMAGE_SIZE,
isTrain=True,
zmin=-1,
zmax=1,
batch=BATCH,
shuffle_read=False):
if isTrain:
augs = [
imgaug.ResizeShortestEdge(int(size * 1.143)),
imgaug.RandomCrop(size),
imgaug.Flip(horiz=True),
]
else:
augs = [
imgaug.ResizeShortestEdge(int(size * 1.143)),
imgaug.CenterCrop(size),
]
def get_images(dir):
files = glob.glob(os.path.join(dir, "*.jpg"))
if shuffle_read:
import random
random.seed(1)
random.shuffle(files)
else:
files = sorted(files)
image_df = ImageFromFile(files, channel=3, shuffle=isTrain)
image_df = AugmentImageComponent(image_df, augs)
random_df = RandomZData([size, size, 3], zmin, zmax)
return JoinData([random_df, image_df])
names = ['train'] if isTrain else ['test']
df = get_images(*[os.path.join(datadir, n) for n in names])
df = BatchData(df, batch)
return df
def get_train_augmentors(self, input_shape, output_shape, view=False):
print(input_shape, output_shape)
if self.model_mode == "class_rotmnist":
shape_augs = [
imgaug.Affine(rotate_max_deg=359,
interp=cv2.INTER_NEAREST,
border=cv2.BORDER_CONSTANT),
]
input_augs = []
else:
shape_augs = [
imgaug.Affine(
rotate_max_deg=359,
translate_frac=(0.01, 0.01),
interp=cv2.INTER_NEAREST,
border=cv2.BORDER_REFLECT,
),
imgaug.Flip(vert=True),
imgaug.Flip(horiz=True),
imgaug.CenterCrop(input_shape),
]
input_augs = [
imgaug.RandomApplyAug(
imgaug.RandomChooseAug([
GaussianBlur(),
MedianBlur(),
imgaug.GaussianNoise(),
]),
0.5,
),
# Standard colour augmentation
imgaug.RandomOrderAug([
imgaug.Hue((-8, 8), rgb=True),
imgaug.Saturation(0.2, rgb=True),
imgaug.Brightness(26, clip=True),
imgaug.Contrast((0.75, 1.25), clip=True),
]),
imgaug.ToUint8(),
]
if self.model_mode == "seg_gland":
label_augs = []
label_augs = [GenInstanceContourMap(mode=self.model_mode)]
label_augs.append(BinarizeLabel())
if not view:
label_augs.append(imgaug.CenterCrop(output_shape))
else:
label_augs.append(imgaug.CenterCrop(input_shape))
return shape_augs, input_augs, label_augs
elif self.model_mode == "seg_nuc":
label_augs = []
label_augs = [GenInstanceMarkerMap()]
label_augs.append(BinarizeLabel())
if not view:
label_augs.append(imgaug.CenterCrop(output_shape))
else:
label_augs.append(imgaug.CenterCrop(input_shape))
return shape_augs, input_augs, label_augs
else:
return shape_augs, input_augs
def fbresnet_augmentor(isTrain, crop_method, color_augmentation):
"""
Augmentor used in fb.resnet.torch, for BGR images in range [0,255].
"""
execution_lst = []
if isTrain:
augmentors = [
# 1. crop_method
# a) GoogleNetResize
GoogleNetResize(),
# b) ShortestEdgeResize
imgaug.ResizeShortestEdge(256),
# c) GlobalWarp
imgaug.Resize(226), # NOTE: for CAM generation
imgaug.RandomCrop((224, 224)),
# d) CAMCrop
# (when CAMCrop is set, the output from the original DataFlow has already been cropped)
# 2. color_augmentation
imgaug.RandomOrderAug([
imgaug.BrightnessScale((0.6, 1.4), clip=False),
imgaug.Contrast((0.6, 1.4), clip=False),
imgaug.Saturation(0.4, rgb=False),
# rgb-bgr conversion for the constants copied from fb.resnet.torch
imgaug.Lighting(
0.1,
eigval=np.asarray([0.2175, 0.0188, 0.0045][::-1]) * 255.0,
eigvec=np.array([[-0.5675, 0.7192, 0.4009],
[-0.5808, -0.0045, -0.8140],
[-0.5836, -0.6948, 0.4203]],
dtype='float32')[::-1, ::-1])
]),
imgaug.Flip(horiz=True),
]
#
if crop_method == 'GoogleNetResize':
print(
'--> perform GoogleNetResize cropping method during the training pipeline'
)
execution_lst.extend([0])
elif crop_method == 'ShortestEdgeResize':
print(
'--> perform ShortestEdgeResize cropping method during the training pipeline'
)
execution_lst.extend([1, 3])
elif crop_method == 'GlobalWarp':
print(
'--> perform GlobalWarp cropping method during the training pipeline'
)
execution_lst.extend([2, 3])
elif crop_method == 'CAMCrop':
# enable CAMCrop @ 20171124
print(
'*** Perform CAMCrop to better the training dynamics and the results ***'
)
if color_augmentation:
print(
'--> perform color augmentation during the training pipeline')
execution_lst.extend([4])
else:
print(
'--> discard the color jittering process during the training pipeline'
)
# perform mirror reflection augmentation anyway
execution_lst.extend([5])
else:
augmentors = [
imgaug.ResizeShortestEdge(256, cv2.INTER_CUBIC),
imgaug.CenterCrop((224, 224)),
imgaug.RandomCrop((224, 224)),
]
if crop_method == 'RandomCrop':
execution_lst.extend([0, 2])
elif crop_method == 'CenterCrop':
execution_lst.extend([0, 1])
return [
item_ for id_, item_ in enumerate(augmentors) if id_ in execution_lst
]
def prediction_incorrect(logits,
label,
topk=1,
name='incorrect_vector'):
with tf.name_scope('prediction_incorrect'):
x = tf.logical_not(tf.nn.in_top_k(logits, label, topk))
return tf.cast(x, tf.float32, name=name)
wrong = prediction_incorrect(logits, label, 1, name='wrong-top1')
add_moving_summary(tf.reduce_mean(wrong, name='train-error-top1'))
wrong = prediction_incorrect(logits, label, 5, name='wrong-top5')
add_moving_summary(tf.reduce_mean(wrong, name='train-error-top5'))
return loss
if __name__ == '__main__':
import argparse
from tensorpack.dataflow import TestDataSpeed
parser = argparse.ArgumentParser()
parser.add_argument('--data', required=True)
parser.add_argument('--batch', type=int, default=32)
args = parser.parse_args()
augs = fbresnet_augmentor(False)
augs = [imgaug.ResizeShortestEdge(256), imgaug.CenterCrop(224)]
df = get_imagenet_dataflow(args.data, 'train', args.batch, augs)
TestDataSpeed(df).start()
def simple_augmentor():
augmentors = [
imgaug.MapImage(lazy_resize),
imgaug.CenterCrop((224, 224)),
]
return augmentors
def get_train_augmentors(self, input_shape, output_shape, view=False):
print(input_shape, output_shape)
shape_augs = [
imgaug.Affine(
shear=5, # in degree
scale=(0.8, 1.2),
rotate_max_deg=179,
translate_frac=(0.01, 0.01),
interp=cv2.INTER_NEAREST,
border=cv2.BORDER_CONSTANT),
imgaug.Flip(vert=True),
imgaug.Flip(horiz=True),
imgaug.CenterCrop(input_shape),
]
input_augs = [
imgaug.RandomApplyAug(
imgaug.RandomChooseAug(
[
GaussianBlur(),
MedianBlur(),
imgaug.GaussianNoise(),
]
), 0.5),
# standard color augmentation
imgaug.RandomOrderAug(
[imgaug.Hue((-8, 8), rgb=True),
imgaug.Saturation(0.2, rgb=True),
imgaug.Brightness(26, clip=True),
imgaug.Contrast((0.75, 1.25), clip=True),
]),
CoarseDropout(),
SaltPepperNoise(),
imgaug.ToUint8(),
]
#input_augs = [
# [imgaug.Hue((-8, 8), rgb=True),
# imgaug.Saturation(0.2, rgb=True),
# imgaug.Brightness(26, clip=True),
# imgaug.Contrast((0.75, 1.25), clip=True),
# ]),
# CoarseDropout(),
# SaltPepperNoise(),
# imgaug.ToUint8(),]
label_augs = []
if self.model_type == 'unet' or self.model_type == 'micronet':
label_augs =[GenInstanceUnetMap(crop_shape=output_shape)]
if self.model_type == 'dcan':
label_augs =[GenInstanceContourMap(crop_shape=output_shape)]
if self.model_type == 'dist':
label_augs = [GenInstanceDistance(crop_shape=output_shape, inst_norm=False)]
if self.model_type == 'np_hv':
label_augs = [GenInstanceHV(crop_shape=output_shape)]
if self.model_type == 'np_dist':
label_augs = [GenInstanceDistance(crop_shape=output_shape, inst_norm=True)]
if not self.type_classification:
label_augs.append(BinarizeLabel())
if not view:
label_augs.append(imgaug.CenterCrop(output_shape))
return shape_augs, input_augs, label_augs