Running scib.metrics.metrics() produces dataframe of NaNs

[EvanDMee]

Hello,

I am attempting to run scib metrics with a merged object and an integrated object both of which were created in Seurat and imported for scib as .h5ad. When I try to run scib.metrics.metrics() with these objects it outputs a dataframe with NaNs for each metric and no error message. What could be going wrong here? I do not see any documentation as to what the function expects for each argument so I cannot determine what I may be doing wrong.

Thanks!

[WhatMelonGua]

hi, I have also been exploring the use of sciB recently. Have you set the correct parameters for the functions 'scib.metrics.metrics()'? such as [hvgscore , nmi, ari ......] you can see more about this function by the website below: https://scib.readthedocs.io/en/latest/api/scib.metrics.metrics.html#scib.metrics.metrics

And during my use, you can set the parameters to True(like xxx_=True) to get that metric. If you don't set them,they are default False,then the function will fill the False metric with NA I don't know if that's the wrong parameter settings makes it happened.

for example,I set nmi [one of the Parameter] to True, then I'll get a dataframe with nmi(a value),and the metrics left are all NA