How to compute the integration metrics

Hi, I want to ask how to compute integration metrics, such as ARI, normalized mutual information, graph connectivity, batch LISI (iLISI), cell-type LISI (cLISI) and so on? Cause I can’t find the related code on GitHub and Zenodo.

I want to if the raw data has the label annotated by experts as the ground truth? For example, best to my knowledge, I think the ARI should be computed by the predicted clusters and the raw data true labels(ground truth), if I just cluster the raw data for original labels and compute it with predicted clusters for ARI, I think the metrics is not convincing right?

For computing integration metrics, you may refer to scib-metrics or the original implementations in scib.