Preprocess for pretrained VGG_bn model

1

Hi,

Now, pretrained VGG13~19 models with batch normalization are provided.
https://github.com/pytorch/vision/blob/master/torchvision/models/vgg.py

Are preprocess for them same as process for models without batch normalization?
(transforms.Normalize(mean=[0.485, 0.456, 0.406],std=[0.229, 0.224, 0.225]))

I would like to confirm since bellowing paper mentioned normalizing for input image was omitted and it make sense because it could be substituted by batch norm layer.
https://arxiv.org/pdf/1612.01452.pdf

How were models provided in the first link trained?

Thanks.

1 Like
2

Yes they are pre-processed in the same way as all other models.

3

Thanks! I have got it.