Why did you use the colorjitter and lighting data augment during training

[leoozy]

Hello, I noticed that you used the color jitter and lighting during training the imagenet. But normally we will not apply these two data augment for a fair comparison. Could you help me with this confusion?
transform_train_pre = transforms.Compose([ transforms.RandomResizedCrop(224, scale=(0.08, 1.0), interpolation=Image.BICUBIC), transforms.RandomHorizontalFlip(), transforms.ColorJitter( brightness=0.4, contrast=0.4, saturation=0.4, ), ]) transform_train_after = transforms.Compose([ transforms.ToTensor(), Lighting(0.1, _IMAGENET_PCA['eigval'], _IMAGENET_PCA['eigvec']), transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) ])