Hello scverse community,
I am using scVI for integrating multiple samples coming from different studies.
For doing this, I am referring to the scVI basic tutorial.
This is what my anndata looks like after feature selection and setting up the training:
When I train the model, however, I get multiple epoch sets, all of them with the same amount of epochs, as follows:
And I end up with as many bars as samples, even if the covariate key for the integration is “Study” only (so I guess the model is not informed about samples). It’s the first time that I see this using scVI. I have always seen one epoch bar with a precise amount of epochs.
Since I am not familiar at all with these techniques, can someone please briefly explain what is happening?