Tim Sainburg
Tim Sainburg
If you provide a specific issue and colab link reproducing it I can take a look. As it stands this issue is too vaguely described.
Thanks for providing the colab notebook. Note that you are plotting the results on the training data here and not the held out testing data. This is very important when...
I updated the notebooks to match the keras update in this PR. Have not pulled it in yet: https://github.com/lmcinnes/umap/pull/1123
We will need to merge this with https://github.com/lmcinnes/umap/pull/1123
Noisereduce iterates over chunks so length doesn’t matter. It looks like the it thinks your data has 268416000 channels which probably means you just need to transpose the input data...
this is related to https://github.com/lmcinnes/umap/pull/1153 maybe @jacobgolding knows what is going on, I don't have much experience with landmarks yet
I can confirm that this is an issue in colab with the most recent updates. https://colab.research.google.com/drive/1QuZrOUDHaOg2VnL-RoeVjLUYsFNJj2Z4 There have been a number of updates from @AMS-Hippo and @jacobgolding - any idea...
I played around a bit with checking out previous commits going back before these updates and the problem persists, so maybe this has to do with incompatibility with new tensorflow...
hey, can you try this branch to see if it resolves the issue on python 3.11? https://github.com/lmcinnes/umap/pull/1123
We are just about to pull in an updated version of Parametric UMAP https://github.com/lmcinnes/umap/pull/1153 so my plan is to wait till that is pulled in to integrate #1123