Open sokratiag opened 1 month ago
I have encountered the same problem, do you have a solution?
this bug was caused by cluster name. I fixed this by:
adata = adata[~adata.obs['clusters'].isin([None,np.nan])].copy() #cluster / louvain
filter adata before the first step
Hello,
I have encountered the following error in scv.tl.paga:
adata.uns['neighbors']['distances'] = adata.obsp['distances'] adata.uns['neighbors']['connectivities'] = adata.obsp['connectivities']
scv.tl.paga(adata, groups='Clusters_new') df = scv.get_df(adata, 'paga/transitions_confidence', precision=2).T df.style.background_gradient(cmap='Blues').format('{:.2g}')
versions: scvelo 0.3.2 (python 3.8.3) igraph 0.11.6
Error:
~/.local/lib/python3.8/site-packages/igraph/clustering.py in cluster_graph(self, combine_vertices, combine_edges) 334 """ 335 result = self.graph.copy() --> 336 result.contract_vertices(self.membership, combine_vertices) 337 if combine_edges is not False: 338 result.simplify(combine_edges=combine_edges)
InternalError: Error at src/constructors/basic_constructors.c:75: Invalid (negative or too large) vertex ID. -- Invalid vertex ID
Any ideas on how to solve it? Thanks!