I recently got the following problem while analyzing CITE-Seq data with TOTALVI:
RuntimeWarning: The lr scheduler dict contains the key(s) ['monitor'], but the keys will be ignored. you need to call lr_scheduler.step()
manually in manual optimization.
I’m not sure what’s wrong, has anyone had a similar problem? Can you provide any help?
My code is as follows:
adata = mdata["rna"].copy() adata.obsm["protein_expression"] = mdata["protein"].layers["counts"].A.copy() scvi.model.TOTALVI.setup_anndata( adata, layer="counts", batch_key="Batch", protein_expression_obsm_key="protein_expression", ) arches_params = dict( use_layer_norm="Both", use_batch_norm="none", n_layers_decoder=2, n_layers_encoder=2, ) vae = scvi.model.TOTALVI(adata, **arches_params) vae.train()
By the way, the versions of scvi and torch are 1.0.2 and 2.0.1+cu117 respectively.