def create_image_mb_source(map_file, mean_file, train, total_number_of_samples):
if not os.path.exists(map_file) or not os.path.exists(mean_file):
raise RuntimeError("File '%s' or '%s' does not exist." % (map_file, mean_file))
# transformation pipeline for the features has jitter/crop only when training
transforms = []
if train:
transforms += [
xforms.crop(crop_type='randomarea', area_ratio=(0.08, 1.0), aspect_ratio=(0.75, 1.3333), jitter_type='uniratio'), # train uses jitter
xforms.color(brightness_radius=0.4, contrast_radius=0.4, saturation_radius=0.4)
]
else:
transforms += [
C.io.transforms.crop(crop_type='center', side_ratio=0.875) # test has no jitter
]
transforms += [
xforms.scale(width=image_width, height=image_height, channels=num_channels, interpolations='cubic'),
xforms.mean(mean_file)
]
# deserializer
return C.io.MinibatchSource(
C.io.ImageDeserializer(map_file, C.io.StreamDefs(
features=C.io.StreamDef(field='image', transforms=transforms), # 1st col in mapfile referred to as 'image'
labels=C.io.StreamDef(field='label', shape=num_classes))), # and second as 'label'
randomize=train,
max_samples=total_number_of_samples,
multithreaded_deserializer=True)
def create_reader(map_file, train, dimensions, classes,
total_number_of_samples):
print(
f"Reading map file: {map_file} with number of samples {total_number_of_samples}"
)
# transformation pipeline for the features has jitter/crop only when training
transforms = []
# finalize_network uses data augmentation (translation only)
if train:
transforms += [
xforms.crop(crop_type='randomside',
area_ratio=(0.08, 1.0),
aspect_ratio=(0.75, 1.3333),
jitter_type='uniratio'),
xforms.color(brightness_radius=0.4,
contrast_radius=0.4,
saturation_radius=0.4)
]
transforms += [
xforms.scale(width=dimensions['width'],
height=dimensions['height'],
channels=dimensions['depth'],
interpolations='linear')
]
source = MinibatchSource(ImageDeserializer(
map_file,
StreamDefs(features=StreamDef(field='image', transforms=transforms),
labels=StreamDef(field='label', shape=len(classes)))),
randomize=train,
max_samples=total_number_of_samples,
multithreaded_deserializer=True)
return source
def create_image_mb_source(map_file, train, total_number_of_samples):
print('Creating source for {}.'.format(map_file))
transforms = []
if train:
transforms += [
xforms.crop(crop_type='randomarea',
area_ratio=(0.08, 1.0),
aspect_ratio=(0.75, 1),
jitter_type='uniratio'), # train uses jitter
xforms.color(brightness_radius=0.4,
contrast_radius=0.4,
saturation_radius=0.4)
]
transforms += [
xforms.scale(width=image_width,
height=image_height,
channels=num_channels,
interpolations='cubic')
]
# deserializer
return C.io.MinibatchSource(
C.io.ImageDeserializer(
map_file,
C.io.StreamDefs(
features=C.io.StreamDef(
field='image', transforms=transforms
), # 1st col in mapfile referred to as 'image'
labels=C.io.StreamDef(
field='label',
shape=num_classes))), # and second as 'label'
randomize=train,
max_samples=total_number_of_samples,
multithreaded_deserializer=True)
def create_image_mb_source(map_file, train, total_number_of_samples):
'''
Input: map_file, train:bool, total_num_of_samples
Function:
- checks if it is training or testing phase
- for training: It applies Image Augmentation techniques like cropping, width_shift, height_shift,
horizontal_flip, color_contrast to prevent overfitting of the model.
Return: MinibatchSource to be fed into the CNTK Model
'''
print('Creating source for {}.'.format(map_file))
transforms = []
if train:
# Apply translational and color transformations only for the Image set
transforms += [
xforms.crop(crop_type='randomarea', area_ratio=(0.08, 1.0), aspect_ratio=(0.75, 1), jitter_type='uniratio'), # train uses jitter
xforms.color(brightness_radius=0.4, contrast_radius=0.4, saturation_radius=0.4)
]
# Scale the images to a specified size (224 x 224) as expected by the model
transforms += [
xforms.scale(width=image_width, height=image_height, channels=num_channels, interpolations='cubic')
]
return C.io.MinibatchSource(
C.io.ImageDeserializer(
map_file,
C.io.StreamDefs(features=C.io.StreamDef(field='image', transforms=transforms), # 1st col in mapfile referred to as 'image'
labels=C.io.StreamDef(field='label', shape=num_classes))), # and second as 'label'
randomize=train,
max_samples=total_number_of_samples,
multithreaded_deserializer=False)
def create_image_mb_source(map_file, mean_file, train, total_number_of_samples):
if not os.path.exists(map_file) or not os.path.exists(mean_file):
raise RuntimeError("File '%s' or '%s' does not exist." % (map_file, mean_file))
# transformation pipeline for the features has jitter/crop only when training
transforms = []
if train:
transforms += [
xforms.crop(crop_type='randomarea', area_ratio=(0.08, 1.0), aspect_ratio=(0.75, 1.3333), jitter_type='uniratio'), # train uses jitter
xforms.color(brightness_radius=0.4, contrast_radius=0.4, saturation_radius=0.4)
]
else:
transforms += [
C.io.transforms.crop(crop_type='center', side_ratio=0.875) # test has no jitter
]
transforms += [
xforms.scale(width=image_width, height=image_height, channels=num_channels, interpolations='cubic'),
xforms.mean(mean_file)
]
# deserializer
return C.io.MinibatchSource(
C.io.ImageDeserializer(map_file, C.io.StreamDefs(
features=C.io.StreamDef(field='image', transforms=transforms), # 1st col in mapfile referred to as 'image'
labels=C.io.StreamDef(field='label', shape=num_classes))), # and second as 'label'
randomize=train,
max_samples=total_number_of_samples,
multithreaded_deserializer=True)
def create_mb(map_file, params, training_set):
transforms = []
image_dimensions = params['image_dimensions']
num_classes = params['num_classes']
if training_set:
# Scale to square-sized image. without this the cropping transform would chop the larger dimension of an
# image to make it squared, and then take 0.9 crops from within the squared image.
transforms += [
xforms.scale(width=2 * image_dimensions[0],
height=2 * image_dimensions[1],
channels=image_dimensions[2],
scale_mode='pad',
pad_value=114)
]
transforms += [
xforms.crop(crop_type='randomside',
side_ratio=0.9,
jitter_type='uniratio')
] # Randomly crop square area
#randomside enables Horizontal flipping
#new_dim = side_ratio * min(old_w,old_h) , 0.9 * 224 = 201.6
#transforms += [xforms.crop(crop_type='center')]
transforms += [
xforms.color(brightness_radius=0.2,
contrast_radius=0.2,
saturation_radius=0.2)
]
transforms += [xforms.crop(crop_type='center',
side_ratio=0.875)] # test has no jitter]
# Scale down and pad
transforms += [
xforms.scale(width=image_dimensions[0],
height=image_dimensions[1],
channels=image_dimensions[2],
scale_mode='pad',
pad_value=114)
]
return MinibatchSource(ImageDeserializer(
map_file,
StreamDefs(features=StreamDef(field='image', transforms=transforms),
labels=StreamDef(field='label', shape=num_classes))),
randomize=training_set,
multithreaded_deserializer=True)
def create_minibatch_source(map_filename, num_classes): transforms = [xforms.crop(crop_type='randomside', side_ratio=0.85, jitter_type='uniratio'), xforms.scale(width=224, height=224, channels=3, interpolations='linear'), xforms.color(brightness_radius=0.2, contrast_radius=0.2, saturation_radius=0.2)] return(cntk.io.MinibatchSource(cntk.io.ImageDeserializer( map_filename, cntk.io.StreamDefs( features=cntk.io.StreamDef( field='image', transforms=transforms, is_sparse=False), labels=cntk.io.StreamDef( field='label', shape=num_classes, is_sparse=False)))))
def create_mb_source(map_file, image_width, image_height, num_channels, num_classes, boTrain):
transforms = []
if boTrain:
# Scale to square-sized image. without this the cropping transform would chop the larger dimension of an
# image to make it squared, and then take 0.9 crops from within the squared image.
transforms += [xforms.scale(width=2*image_width, height=2*image_height, channels=num_channels,
interpolations='linear', scale_mode='pad', pad_value=114)]
transforms += [xforms.crop(crop_type='randomside', side_ratio=0.9, jitter_type='uniratio')] # Randomly crop square area
transforms += [xforms.scale(width=image_width, height=image_height, channels=num_channels, # Scale down and pad
interpolations='linear', scale_mode='pad', pad_value=114)]
if boTrain:
transforms += [xforms.color(brightness_radius=0.2, contrast_radius=0.2, saturation_radius=0.2)]
return MinibatchSource(ImageDeserializer(map_file, StreamDefs(
features = StreamDef(field='image', transforms=transforms),
labels = StreamDef(field='label', shape=num_classes))),
randomize = boTrain,
multithreaded_deserializer=True)
def create_reader(map_file, is_train):
transforms = [
xforms.color(brightness_radius=0.2,
contrast_radius=0.2,
saturation_radius=0.2),
xforms.scale(width=img_width,
height=img_height,
channels=img_channel,
interpolations="linear")
]
return C.io.MinibatchSource(
C.io.ImageDeserializer(
map_file,
C.io.StreamDefs(image=C.io.StreamDef(field="image",
transforms=transforms),
dummy=C.io.StreamDef(field="label",
shape=num_classes))),
randomize=is_train,
max_sweeps=C.io.INFINITELY_REPEAT if is_train else 1)
def create_reader(map_file):
transforms = [
xforms.crop(crop_type='randomside',
side_ratio=0.85,
jitter_type='uniratio'),
xforms.scale(width=image_width,
height=image_height,
channels=num_channels,
interpolations='linear'),
xforms.color(brightness_radius=0.2,
contrast_radius=0.2,
saturation_radius=0.2)
]
return (MinibatchSource(
ImageDeserializer(
map_file,
StreamDefs(features=StreamDef(field='image',
transforms=transforms,
is_sparse=False),
labels=StreamDef(field='label',
shape=num_classes,
is_sparse=False)))))