How to train Inceptionv3 model with high-resolution images (ex. 512x512, 2048x2048)?

I’d like to train the Inceptionv3 model with high-resolution images like 512x512. As I know, the Inceptionv3 is designed to work with variable size of images. However, if I train it with images which is not 299x299 size, it return with errors like: size mismatch, m1: [8 x 19200], m2: [768 x 1000]

It’s looks like have some problem with this line.

if self.training and self.aux_logits: aux = self.AuxLogits(x)`

https://github.com/pytorch/vision/blob/62cbf0bfd73c3a258e722cd3b468269c5b07b579/torchvision/models/inception.py#L108

Anybody can help me?

Thanks a lot in advance

[ekagra-ranjan]

I guess this can be closed due to #744.

[fmassa]

Hi,

The issue is that we have a fixed avg_pool2d in https://github.com/pytorch/vision/blob/master/torchvision/models/inception.py#L117 instead of a global avg pooling.

The solution for now is to copy-paste the implementation and modify that line to use a F.adaptive_avg_pool2d(x, (1, 1)).