Chuan Xu

@anemartinezlarrinaga2898, I will close this issue for now. If you still have problems, please let us know or feel free to reopen it.

@hxpGit512, you can only compare scores under one model but not across models. In cases of “unassigned”or “heterogeneous”, the score of a cell represents the probability when predicting this cell...

I will close the issue as this should be solved. Please let us know if you still have problems.

@akk01, this error means you cannot fetch `https://celltypist.cog.sanger.ac.uk/models/models.json` within 30 seconds of response. Have you solved it now?

@saeedfc, good point! The 'Cells_Intestinal_Tract' indeed contains pediatric CD cells. Will clarify this in the model details.

@saeedfc, the model details have been updated. https://www.celltypist.org/models https://github.com/Teichlab/celltypist_wiki/blob/main/json/models.json#L56

This issue should be solved. Please let me know if you still need more details for the model.

@markdane, if you want to use a csv file as the input, a raw count matrix (that only contains integers) is required. If you use an AnnData as input, you...

@colinmegill , I think one of the solutions at the user end (I haven't tried myself though..) is to turn the x and y axis into [0,1], something like `umaps...

@dkapadia612, you can train the model using any numbers of genes. There is no definitive relationship between the accuracy of the model and the number of genes (for example, a...