Fabrizio Ottati

I managed to use datapipes for the decompression (no code to maintain for us, yeeee :laughing:) but between the old and new dataset, on NMNIST, there is a 30% slowdown...

I onestly do not know :disappointed: If `keep_compressed=False`, the dataset in uncompressed in a dedicated folder at object creation, in the following way: ```python def _uncompress(self, dp): dp = FileLister(path_to_zip)...

I think the code is decent now. Let's merge like there is no tomorrow :laughing:

> One more thing, is there a reason we're not using this anymore > > https://github.com/neuromorphs/tonic/blob/beef9808639693762893b9f17106a6d9ea61bc74/tonic/io.py#L157 Yes, the reason is that `read_mnist_file` wants the file in input, while the datapipe...

Have you tried to open directly the link of the dataset?

Then the link is broken and the Tonic code has nothing to do with it.

Hi! Sorry to hear that but, as you can imagine, we do not check plagiarism in papers 🤣 The code was added by the first author of the paper, as...

On a second thought, this matter is too sensitive for us to decide. Either get in touch with the other paper's author and reach an agreement, or we wait for...

I think you should an email to the authors to clarify the situation! Maybe they have a new link. On 9 Apr 2024, at 20:43, Jan Finkbeiner ***@***.***> wrote: Hi...

Usually with the drive stuff it is better if the user does it manually like with Prophesee stuff?