Maren Büttner

Yes, I thought so, too, because the fit step works just fine. Thanks for linking the issue to the PyTorch lightning repo, too.

Hi there, I have got a reply from the PyTorch lightning developers, who suggested some change in the DataLoader, see [issue posted there](https://github.com/PyTorchLightning/pytorch-lightning/issues/12917). Is this something to implement without much...

Hi, thank you very much for sharing your thoughts! I agree that you may use kBET for other data types, too. The only limit is that you should have a...

Hi @elva0328 thank you for using kBET! About your questions: 1. kBET is a very sensitive method and will almost always report the presence of a batch effect. In general,...

Hi @elva0328 thanks for the very detailed issue description. I think that you can adjust some of the parameters in kBET to accomodate for your input. As you are already...

Hi, can you please elaborate your issue? E.g. what kind of data do you have, what you would like to sample, what you would like to show and what you...

Hi @yycunc thank you for posting your issue and I am sorry for my late reply. To date, kBET accepts only input in the dense matrix format or a knn-graph...

Hi csennis, I'm sorry to hear about the bad performance of the knn search. I cannot comment on the optimal neighbourhood size because this depends on the number of batches....

Hi @LeonHafner, thank you for your request. I am not quite sure what kind of structure you would like to use to compute kBET, though. I assume that you want...

Hi, if this is still of interest: I recommend to reduce the number of features (genes) through computing PCA/SVD first and run the knn-graph on the low-dimensional object instead. Best,...