super-gradients/super_gradients/training/datasets/auto_augment.py

""" RandAugment
RandAugment is a variant of AutoAugment which randomly selects transformations
 from AutoAugment to be applied on an image.

RandomAugmentation Implementation adapted from:
    https://github.com/rwightman/pytorch-image-models/blob/master/timm/data/auto_augment.py

Papers:
    RandAugment: Practical automated data augmentation... - https://arxiv.org/abs/1909.13719

"""
import random
import re
from PIL import Image, ImageOps, ImageEnhance
import numpy as np


_FILL = (128, 128, 128)

# to unify the calls of the different augmentations in terms of params, all augmentations are set to work with a single
# magnitude params, normalized according to _MAX_MAGNITUDE
_MAX_MAGNITUDE = 10.

_HPARAMS_DEFAULT = dict(
    translate_const=250,
    img_mean=_FILL,
)

# Define the interpolation types
_RANDOM_INTERPOLATION = Image.BILINEAR


def _interpolation(kwargs):
    """
    Performs Bi-Linear interpolation
    """
    interpolation = kwargs.pop('resample', Image.BILINEAR)
    if isinstance(interpolation, (list, tuple)):
        return random.choice(interpolation)
    else:
        return interpolation


def shear_x(img, factor, **kwargs):
    return img.transform(img.size, Image.AFFINE, (1, factor, 0, 0, 1, 0), **kwargs)


def shear_y(img, factor, **kwargs):
    return img.transform(img.size, Image.AFFINE, (1, 0, 0, factor, 1, 0), **kwargs)


def translate_x_rel(img, pct, **kwargs):
    pixels = pct * img.size[0]
    return img.transform(img.size, Image.AFFINE, (1, 0, pixels, 0, 1, 0), **kwargs)


def translate_y_rel(img, pct, **kwargs):
    pixels = pct * img.size[1]
    return img.transform(img.size, Image.AFFINE, (1, 0, 0, 0, 1, pixels), **kwargs)


def translate_x_abs(img, pixels, **kwargs):
    return img.transform(img.size, Image.AFFINE, (1, 0, pixels, 0, 1, 0), **kwargs)


def translate_y_abs(img, pixels, **kwargs):
    return img.transform(img.size, Image.AFFINE, (1, 0, 0, 0, 1, pixels), **kwargs)


def rotate(img, degrees, **kwargs):
    return img.rotate(degrees, **kwargs)


def auto_contrast(img, **__):
    return ImageOps.autocontrast(img)


def invert(img, **__):
    return ImageOps.invert(img)


def equalize(img, **__):
    return ImageOps.equalize(img)


def solarize(img, thresh, **__):
    return ImageOps.solarize(img, thresh)


def solarize_add(img, add, thresh=128, **__):
    lut = []
    for i in range(256):
        if i < thresh:
            lut.append(min(255, i + add))
        else:
            lut.append(i)
    if img.mode in ("L", "RGB"):
        if img.mode == "RGB" and len(lut) == 256:
            lut = lut + lut + lut
        return img.point(lut)
    else:
        return img


def posterize(img, bits_to_keep, **__):
    if bits_to_keep >= 8:
        return img
    return ImageOps.posterize(img, bits_to_keep)


def contrast(img, factor, **__):
    return ImageEnhance.Contrast(img).enhance(factor)


def color(img, factor, **__):
    return ImageEnhance.Color(img).enhance(factor)


def brightness(img, factor, **__):
    return ImageEnhance.Brightness(img).enhance(factor)


def sharpness(img, factor, **__):
    return ImageEnhance.Sharpness(img).enhance(factor)


def _randomly_negate(v):
    """With 50% prob, negate the value"""
    return -v if random.random() > 0.5 else v


def _rotate_level_to_arg(level, _hparams):
    # range [-30, 30]
    level = (level / _MAX_MAGNITUDE) * 30.
    level = _randomly_negate(level)
    return level,


def _enhance_level_to_arg(level, _hparams):
    # range [0.1, 1.9]
    return (level / _MAX_MAGNITUDE) * 1.8 + 0.1,


def _enhance_increasing_level_to_arg(level, _hparams):
    # range [0.1, 1.9]
    level = (level / _MAX_MAGNITUDE) * .9
    level = 1.0 + _randomly_negate(level)
    return level,


def _shear_level_to_arg(level, _hparams):
    # range [-0.3, 0.3]
    level = (level / _MAX_MAGNITUDE) * 0.3
    level = _randomly_negate(level)
    return level,


def _translate_abs_level_to_arg(level, hparams):
    translate_const = hparams['translate_const']
    level = (level / _MAX_MAGNITUDE) * float(translate_const)
    level = _randomly_negate(level)
    return level,


def _translate_rel_level_to_arg(level, hparams):
    # default range [-0.45, 0.45]
    translate_pct = hparams.get('translate_pct', 0.45)
    level = (level / _MAX_MAGNITUDE) * translate_pct
    level = _randomly_negate(level)
    return level,


def _posterize_level_to_arg(level, _hparams):
    # As per Tensorflow TPU EfficientNet impl
    # range [0, 4], 'keep 0 up to 4 MSB of original image'
    # intensity/severity of augmentation decreases with level
    return int((level / _MAX_MAGNITUDE) * 4),


def _posterize_increasing_level_to_arg(level, hparams):
    # As per Tensorflow models research and UDA impl
    # range [4, 0], 'keep 4 down to 0 MSB of original image',
    # intensity/severity of augmentation increases with level
    return 4 - _posterize_level_to_arg(level, hparams)[0],


def _posterize_original_level_to_arg(level, _hparams):
    # As per original AutoAugment paper description
    # range [4, 8], 'keep 4 up to 8 MSB of image'
    # intensity/severity of augmentation decreases with level
    return int((level / _MAX_MAGNITUDE) * 4) + 4,


def _solarize_level_to_arg(level, _hparams):
    # range [0, 256]
    # intensity/severity of augmentation decreases with level
    return int((level / _MAX_MAGNITUDE) * 256),


def _solarize_increasing_level_to_arg(level, _hparams):
    # range [0, 256]
    # intensity/severity of augmentation increases with level
    return 256 - _solarize_level_to_arg(level, _hparams)[0],


def _solarize_add_level_to_arg(level, _hparams):
    # range [0, 110]
    return int((level / _MAX_MAGNITUDE) * 110),


LEVEL_TO_ARG = {
    'AutoContrast': None,
    'Equalize': None,
    'Invert': None,
    'Rotate': _rotate_level_to_arg,
    # There are several variations of the posterize level scaling in various Tensorflow/Google repositories/papers
    'Posterize': _posterize_level_to_arg,
    'PosterizeIncreasing': _posterize_increasing_level_to_arg,
    'PosterizeOriginal': _posterize_original_level_to_arg,
    'Solarize': _solarize_level_to_arg,
    'SolarizeIncreasing': _solarize_increasing_level_to_arg,
    'SolarizeAdd': _solarize_add_level_to_arg,
    'Color': _enhance_level_to_arg,
    'ColorIncreasing': _enhance_increasing_level_to_arg,
    'Contrast': _enhance_level_to_arg,
    'ContrastIncreasing': _enhance_increasing_level_to_arg,
    'Brightness': _enhance_level_to_arg,
    'BrightnessIncreasing': _enhance_increasing_level_to_arg,
    'Sharpness': _enhance_level_to_arg,
    'SharpnessIncreasing': _enhance_increasing_level_to_arg,
    'ShearX': _shear_level_to_arg,
    'ShearY': _shear_level_to_arg,
    'TranslateX': _translate_abs_level_to_arg,
    'TranslateY': _translate_abs_level_to_arg,
    'TranslateXRel': _translate_rel_level_to_arg,
    'TranslateYRel': _translate_rel_level_to_arg,
}


NAME_TO_OP = {
    'AutoContrast': auto_contrast,
    'Equalize': equalize,
    'Invert': invert,
    'Rotate': rotate,
    'Posterize': posterize,
    'PosterizeIncreasing': posterize,
    'PosterizeOriginal': posterize,
    'Solarize': solarize,
    'SolarizeIncreasing': solarize,
    'SolarizeAdd': solarize_add,
    'Color': color,
    'ColorIncreasing': color,
    'Contrast': contrast,
    'ContrastIncreasing': contrast,
    'Brightness': brightness,
    'BrightnessIncreasing': brightness,
    'Sharpness': sharpness,
    'SharpnessIncreasing': sharpness,
    'ShearX': shear_x,
    'ShearY': shear_y,
    'TranslateX': translate_x_abs,
    'TranslateY': translate_y_abs,
    'TranslateXRel': translate_x_rel,
    'TranslateYRel': translate_y_rel,
}


class AugmentOp:
    """
    single auto augment operations
    """
    def __init__(self, name, prob=0.5, magnitude=10, hparams=None):
        hparams = hparams or _HPARAMS_DEFAULT
        self.aug_fn = NAME_TO_OP[name]
        self.level_fn = LEVEL_TO_ARG[name]
        self.prob = prob
        self.magnitude = magnitude
        self.hparams = hparams.copy()
        self.kwargs = dict(
            fillcolor=hparams['img_mean'] if 'img_mean' in hparams else _FILL,
            resample=hparams['interpolation'] if 'interpolation' in hparams else _RANDOM_INTERPOLATION,
        )

        # If magnitude_std is > 0, introduce some randomness
        self.magnitude_std = self.hparams.get('magnitude_std', 0)

    def __call__(self, img):
        if self.prob < 1.0 and random.random() > self.prob:
            return img
        magnitude = self.magnitude
        if self.magnitude_std:
            if self.magnitude_std == float('inf'):
                magnitude = random.uniform(0, magnitude)
            elif self.magnitude_std > 0:
                magnitude = random.gauss(magnitude, self.magnitude_std)
        magnitude = min(_MAX_MAGNITUDE, max(0, magnitude))  # clip to valid range
        level_args = self.level_fn(magnitude, self.hparams) if self.level_fn is not None else tuple()
        return self.aug_fn(img, *level_args, **self.kwargs)


_RAND_TRANSFORMS = [
    'AutoContrast',
    'Equalize',
    'Invert',
    'Rotate',
    'Posterize',
    'Solarize',
    'SolarizeAdd',
    'Color',
    'Contrast',
    'Brightness',
    'Sharpness',
    'ShearX',
    'ShearY',
    'TranslateXRel',
    'TranslateYRel',
    # 'Cutout'  # NOTE I've implement this as random erasing separately
]


_RAND_INCREASING_TRANSFORMS = [
    'AutoContrast',
    'Equalize',
    'Invert',
    'Rotate',
    'PosterizeIncreasing',
    'SolarizeIncreasing',
    'SolarizeAdd',
    'ColorIncreasing',
    'ContrastIncreasing',
    'BrightnessIncreasing',
    'SharpnessIncreasing',
    'ShearX',
    'ShearY',
    'TranslateXRel',
    'TranslateYRel',
    # 'Cutout'  # NOTE I've implement this as random erasing separately
]

# These experimental weights are based loosely on the relative improvements mentioned in paper.
# They may not result in increased performance, but could likely be tuned to so.
_RAND_CHOICE_WEIGHTS_0 = {
    'Rotate': 0.3,
    'ShearX': 0.2,
    'ShearY': 0.2,
    'TranslateXRel': 0.1,
    'TranslateYRel': 0.1,
    'Color': .025,
    'Sharpness': 0.025,
    'AutoContrast': 0.025,
    'Solarize': .005,
    'SolarizeAdd': .005,
    'Contrast': .005,
    'Brightness': .005,
    'Equalize': .005,
    'Posterize': 0,
    'Invert': 0,
}


def _select_rand_weights(weight_idx=0, transforms=None):
    transforms = transforms or _RAND_TRANSFORMS
    assert weight_idx == 0  # only one set of weights currently
    rand_weights = _RAND_CHOICE_WEIGHTS_0
    probs = [rand_weights[k] for k in transforms]
    probs /= np.sum(probs)
    return probs


def rand_augment_ops(magnitude=10, hparams=None, transforms=None):
    hparams = hparams or _HPARAMS_DEFAULT
    transforms = transforms or _RAND_TRANSFORMS
    return [AugmentOp(
        name, prob=0.5, magnitude=magnitude, hparams=hparams) for name in transforms]


class RandAugment:
    """
    Random auto augment class, will select auto augment transforms according to probability weights for each op
    """
    def __init__(self, ops, num_layers=2, choice_weights=None):
        self.ops = ops
        self.num_layers = num_layers
        self.choice_weights = choice_weights

    def __call__(self, img):
        # no replacement when using weighted choice
        ops = np.random.choice(
            self.ops, self.num_layers, replace=self.choice_weights is None, p=self.choice_weights)
        for op in ops:
            img = op(img)
        return img


def rand_augment_transform(config_str, hparams):
    """
    Create a RandAugment transform

    :param config_str: String defining configuration of random augmentation. Consists of multiple sections separated by
    dashes ('-'). The first section defines the specific variant of rand augment (currently only 'rand'). The remaining
    sections, not order sepecific determine
        'm' - integer magnitude of rand augment
        'n' - integer num layers (number of transform ops selected per image)
        'w' - integer probabiliy weight index (index of a set of weights to influence choice of op)
        'mstd' -  float std deviation of magnitude noise applied
        'inc' - integer (bool), use augmentations that increase in severity with magnitude (default: 0)
    Ex 'rand-m9-n3-mstd0.5' results in RandAugment with magnitude 9, num_layers 3, magnitude_std 0.5
    'rand-mstd1-w0' results in magnitude_std 1.0, weights 0, default magnitude of 10 and num_layers 2

    :param hparams: Other hparams (kwargs) for the RandAugmentation scheme

    :return: A PyTorch compatible Transform
    """
    magnitude = _MAX_MAGNITUDE  # default to _MAX_MAGNITUDE for magnitude (currently 10)
    num_layers = 2  # default to 2 ops per image
    weight_idx = None  # default to no probability weights for op choice
    transforms = _RAND_TRANSFORMS
    config = config_str.split('-')
    for c in config:
        cs = re.split(r'(\d.*)', c)
        if len(cs) < 2:
            continue
        key, val = cs[:2]
        if key == 'mstd':
            # noise param injected via hparams for now
            hparams.setdefault('magnitude_std', float(val))
        elif key == 'inc':
            if bool(val):
                transforms = _RAND_INCREASING_TRANSFORMS
        elif key == 'm':
            magnitude = int(val)
        elif key == 'n':
            num_layers = int(val)
        elif key == 'w':
            weight_idx = int(val)
        else:
            assert False, 'Unknown RandAugment config section'
    ra_ops = rand_augment_ops(magnitude=magnitude, hparams=hparams, transforms=transforms)
    choice_weights = None if weight_idx is None else _select_rand_weights(weight_idx)
    return RandAugment(ra_ops, num_layers, choice_weights=choice_weights)