Zaccharie Ramzi

For those interested by this problem, I have a fork where I implemented this idea, I have been using it and it's working great so far. I opened a PR,...

For anyone looking for a maintained official implementation, it is available in [tensorflow/models](https://github.com/tensorflow/models/blob/master/official/vision/modeling/backbones/resnet.py). There is however a slight "bug" in the implementation because it has an extra convolution compared to...

The model is now in keras-cv after [this PR](https://github.com/keras-team/keras-cv/pull/805#event-7441560537).

In a different setup, using a different implementation of the same model (ResNet-18 in CIFAR configuration), and a different code to perform the optimization, I also find that without using...

@skrbnv the differences pointed out [in this comment](https://github.com/kuangliu/pytorch-cifar/issues/136#issuecomment-862626209) are just the differences between the CIFAR-10 and the ImageNet versions. In [the original ResNet paper](https://arxiv.org/pdf/1512.03385.pdf), there were indeed 2 different versions...

@mathurinm followed your style for my fig (you can see it in the overleaf), and adapted a bit for example the linewidth and markersize (just because I have much more...

I think it's because `benchopt --version --check-editable` returns: ``` /home/zaccharie/workspace/benchOpt/benchopt/config.py:75: UserWarning: BenchOpt config file /home/zaccharie/.config/benchopt.ini is with mode 664. As it stores sensitive information such as the github token, it...

Yes `chmod 600 ~/.config/benchopt.ini` solved the issue. Keeping this open, since I think it should be better handled.

@tomMoral why is the size a concern? @agramfort no not exactly, more what Thomas is describing.

@Thenerdstation do you know if those ready features of Keras were ported to tf.keras in Tf 2.x? I have been looking around a bit in [here](https://www.tensorflow.org/api_docs/python/tf/test) but didn't find anything...