Gökçen Eraslan

Wait, now I updated jsonschema to 4.20.0 and I cannot reproduce the error. I really didn't change anything in between...

Is there some documentation on what is expected of the custom dataloader's collate function? I can imagine a dict with keys like `X`, `batch` and `labels` just by following up...

It'd be really nice to save labels as tsv in this repo, I fully agree. But for now, one option is to use the file in scanpy tutorial [here](https://github.com/theislab/scanpy_usage/blob/master/170503_zheng17/data/zheng17_bulk_lables.txt), and...

That's because the cell order is different:  Here is the full file with barcodes and labels as tsv: [zheng17-cell-labels.txt](https://github.com/10XGenomics/single-cell-3prime-paper/files/2276933/zheng17-cell-labels.txt)

Barcode order in scanpy file follows the barcode order in http://cf.10xgenomics.com/samples/cell-exp/1.1.0/fresh_68k_pbmc_donor_a/fresh_68k_pbmc_donor_a_filtered_gene_bc_matrices.tar.gz file, fyi. Labels are not from denovo clustering, it's based on correlation with 11 purified bulk samples, same as...

I have something that might be related: ```python ad = ad[ad.obs['cell type'] != 'nan'].copy() assert np.all(ad.obs['cell type'] != 'nan') sc.utils.sanitize_anndata(ad) assert np.all(ad.obs['cell type'] != 'nan') ``` This fails in the...

Yes there are, and this is how I realized it. I saw them in the plots and wondered why they show up after removing them.

Oh you mean real NaNs, no there is not.

Do you get an exception message or something else? If you can also copy paste the error message here, we can debug it more easily.