Massimiliano Patacchiola

I just run the code on my machine multiple times and it worked without any issue. I am using the same configuration used in the original experiments of the paper,...

Hi @g-benton Yes we are aware of this problem which has been already discussed in another [issue](https://github.com/BayesWatch/deep-kernel-transfer/issues/8). Unfortunately, the original links have been removed and we are not aware of...

Hi @ZohrehAdabi Can it be a problem related to the batch-norm statistics? Since the backbone uses batch-norm, you should carefully use the calls to `model.train()` and `model.eval()` to avoid any...

Hi @ZohrehAdabi Things I would check are the following: - Normalization of the images, are you using the same normalization in both functions? - Dataloader, is it possible that the...

@ZohrehAdabi I am not sure where the problem can be. It can be an issue of the dataloader. You could try to comment-out the data manager lines and pass synthetic...

The data_manager seems to be the issue for the previous problem. For the second problem with the two models, are you loading the same pretrained models for both objects or...

Thank you for the proposal. The test file looks good but its location is not appropriate. You placed the test file and the images under the deepgaze folder, which contains...

Hi @sambatra . Can you please explain better what you mean by "original" and "loaded" model? Is the original model the result in the paper and loaded the one you...