X_SSL_Net/ref/fft_wave_pro/MobileMamba/data/transforms.py

import torch
import torch.nn as nn
import torchvision.transforms.functional as F
from torchvision import transforms
from timm.data import create_transform

import cv2
import numpy as np
from PIL import Image
from . import TRANSFORMS


# for torchvision
tv_tran = ["Compose", "ToTensor", "PILToTensor", "ConvertImageDtype", "ToPILImage", "Normalize", "Resize", "Scale",
           "CenterCrop", "Pad", "Lambda", "RandomApply", "RandomChoice", "RandomOrder", "RandomCrop",
           "RandomHorizontalFlip", "RandomVerticalFlip", "RandomResizedCrop", "RandomSizedCrop", "FiveCrop", "TenCrop",
           "LinearTransformation", "ColorJitter", "RandomRotation", "RandomAffine", "Grayscale", "RandomGrayscale",
           "RandomPerspective", "RandomErasing", "GaussianBlur", "InterpolationMode", "RandomInvert", "RandomPosterize",
           "RandomSolarize", "RandomAdjustSharpness", "RandomAutocontrast", "RandomEqualize"]

for tv_tran_name in tv_tran:
	tv_transform = getattr(transforms, tv_tran_name, None)
	TRANSFORMS.register_module(tv_transform, name=tv_tran_name) if tv_transform else None

# for timm
TRANSFORMS.register_module(create_transform, name='timm_create_transform')


class vt_TransBase(object):

	def __init__(self):
		pass

	def pre_process(self):
		pass

	def __call__(self, img):
		pass


@TRANSFORMS.register_module
class vt_Identity(vt_TransBase):

	def __call__(self, img):
		return img


@TRANSFORMS.register_module
class vt_Resize(vt_TransBase):
	"""
	Args:
		size    : h | (h, w)
		img     : PIL Image
	Returns:
		PIL Image
	"""
	def __init__(self, size, interpolation=F.InterpolationMode.BICUBIC):
		super().__init__()
		self.size = size
		self.interpolation = interpolation

	def __call__(self, img):
		return F.resize(img, self.size, self.interpolation)


@TRANSFORMS.register_module
class vt_Compose(vt_TransBase):
	def __init__(self, transforms):
		super().__init__()
		self.transforms = transforms

	def pre_process(self):
		for t in self.transforms:
			t.pre_process()

	def __call__(self, img):
		for t in self.transforms:
			img = t(img)
		return img


if __name__ == '__main__':
	import matplotlib.pyplot as plt
	from skimage import color
	import torch.nn.functional as F1

	path = '../ttt/ttt.png'
	img = Image.open(path).convert('RGB')
	train_transforms = list()
	train_transforms.append(Resize((200, 300)))
	train_transforms.append(Flip(p=0.5, flipCode=1))
	train_transforms.append(ToTensor())
	train_transforms.append(Normalize(mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5)))
	train_transforms = Compose(train_transforms)
	img1 = train_transforms(img)
	print(img1.min(), img1.max())