Zaccharie Ramzi

@fchollet @qlzh727 Ok great! I am currently in the process of porting PyTorch weights to Keras, and I am running into a bit of an issue regarding strided convolutions. I...

@fchollet indeed it somehow had to do with the padding, but more precisely with which side you begin the striding on. Therefore, to compensate the difference between the 2 frameworks,...

@qlzh727 These are the results I obtain with my port of the PyTorch weights on the ImageNet validation set: | **ResNet size** | **Top-1 acc** | **Top-5 acc** | |-----------------|---------------|---------------|...

Do you mean overhead compared to PyTorch or compared to a padding within the convolution op? Anyway I can test both, but for GPU I'll have to check if it...

@fchollet here are the overhead results on just comparing with and without padding in Keras: | **ResNet size / hardware** | **No padding** | **With padding** | |----------------------------|----------------|------------------| | **18-CPU**...

Ok I'll just add the flag before merging. I have to say I hadn't thought about the possibility to use TPUs on Colab. It's true that it might be enough....

Thanks so much for the offer. I am going to try to get my hands dirty by myself a bit first, then I might come back asking for your help...

@qlzh727 you can find the weights ported from PyTorch [in this WeTransfer link](https://we.tl/t-Fo1N1nGEqe). And the evaluation is [in this comment](https://github.com/keras-team/keras/pull/16363#issuecomment-1099193041). @fchollet I added the manual padding flag, so after this...

@qlzh727 I am not sure what you mean by attaching the weights to the PR. I cannot for example put them in a comment. I think it's because it's too...

@qlzh727 Ok I think I did what you asked by creating 2 commits: [one that adds the weights](https://github.com/keras-team/keras/pull/16363/commits/fbeb86bc36337ba612a417e8f5c5b1a5909c9adc) and [one that removes them](https://github.com/keras-team/keras/pull/16363/commits/fb89c6b9af81b56c2bbf5702e063c8d81f79f179). Tell me if it doesn't work for...