brails.processors.foundation_classifier.lib.custom_transforms module

class brails.processors.foundation_classifier.lib.custom_transforms.AddGaussianNoise(mean, sigma, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Add gaussian noise to a numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.AddGaussianPoissonNoise(sigma, peak, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Add poisson noise with gaussian blurred image to a numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.AddSpeckleNoise(mean, sigma, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Add speckle noise to a numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.BilinearResize(zoom)

Bases: object

Resize a PIL.Image or numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.CenterCropNumpy(size)

Bases: object

Crops the given numpy array at the center to have a region of the given size. size can be a tuple (target_height, target_width) or an integer, in which case the target will be of a square shape (size, size)

class brails.processors.foundation_classifier.lib.custom_transforms.ElasticTransform(alpha, sigma)

Bases: object

Apply elastic transformation on a numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.EnhancedCompose(transforms)

Bases: object

Composes several transforms together. :param transforms: list of transforms to compose. :type transforms: List[Transform]

Example

>>> transforms.Compose([
>>>     transforms.CenterCrop(10),
>>>     transforms.ToTensor(),
>>> ])

class brails.processors.foundation_classifier.lib.custom_transforms.MaxScaleNumpy(range_min=0.0, range_max=1.0)

Bases: object

scale with max and min of each channel of the numpy array i.e. channel = (channel - mean) / std

class brails.processors.foundation_classifier.lib.custom_transforms.MedianScaleNumpy(range_min=0.0, range_max=1.0)

Bases: object

Scale with median and mean of each channel of the numpy array i.e. channel = (channel - mean) / std

class brails.processors.foundation_classifier.lib.custom_transforms.Merge(axis=-1)

Bases: object

Merge a group of images

class brails.processors.foundation_classifier.lib.custom_transforms.MutualExclude(exclude_channel, from_channel)

Bases: object

Remove elements from one channel

class brails.processors.foundation_classifier.lib.custom_transforms.NormalizeNumpy

Bases: object

Normalize each channel of the numpy array i.e. channel = (channel - mean) / std

class brails.processors.foundation_classifier.lib.custom_transforms.PoissonSubsampling(peak, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Poisson subsampling on a numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.RandomCropNumpy(size, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Crops the given numpy array at a random location to have a region of the given size. size can be a tuple (target_height, target_width) or an integer, in which case the target will be of a square shape (size, size)

class brails.processors.foundation_classifier.lib.custom_transforms.RandomGaussianBlurring(sigma, p=0.2, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Apply gaussian blur to a numpy.ndarray (H x W x C)

class brails.processors.foundation_classifier.lib.custom_transforms.RandomRotate(angle_range=(0.0, 360.0), axes=(0, 1), mode='reflect', random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

Bases: object

Rotate a PIL.Image or numpy.ndarray (H x W x C) randomly

class brails.processors.foundation_classifier.lib.custom_transforms.Split(*slices, **kwargs)

Bases: object

Split images into individual arrays

brails.processors.foundation_classifier.lib.custom_transforms.center_crop(x, center_crop_size)

brails.processors.foundation_classifier.lib.custom_transforms.elastic_transform(image, alpha=1000, sigma=30, spline_order=1, mode='nearest', random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>): Elastic deformation of image as described in [Simard2003]. .. [Simard2003] Simard, Steinkraus and Platt, “Best Practices for

Convolutional Neural Networks applied to Visual Document Analysis”, in Proc. of the International Conference on Document Analysis and Recognition, 2003.

brails.processors.foundation_classifier.lib.custom_transforms.poisson_downsampling(image, peak, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

brails.processors.foundation_classifier.lib.custom_transforms.random_num_generator(config, random_state=<module 'numpy.random' from '/home/bacetiner/anaconda3/envs/brails-pip/lib/python3.10/site-packages/numpy/random/__init__.py'>)

brails.processors.foundation_classifier.lib.custom_transforms.to_tensor(x)