Closed cristalliao closed 1 year ago
Hi Cristal,
I found a similar error in leiden, caused by different versions of python-igraph package, see https://github.com/atarashansky/SAMap/issues/93. Could you try this solution to see whether it works or not?
Hi Geifei,
I tried this many times, but it still does not work for my server computer. Is there any solution for me to solve this problem?
The error message:
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. leidenalg 0.9.1 requires igraph<0.11,>=0.10.0, but you have igraph 0.9.9 which is incompatible.
Thanks in advance! Very appreciated!
Best regards,
Cristal
Hi Cristal,
It seems that the leigenalg package version 0.9.1 is incompatible with igraph 0.9.9. You can try to also downgrade leigenalg. We have tested STitch3D using python-igraph==0.9.1, louvain==0.7.0, leidenalg==0.7.0, which do not have the issue you reported.
Best, Gefei
Dear Gefei, Thanks a lot! This problem has been solved! Very appreciated! Best regards, Cristal
Dear Professor, I want to know what is the problem with Example 2 in "Visualizing results in 2D". I encountered several problems when I try this great STitch3D in Example 2.
sc.pp.neighbors(model.adata_st, use_rep='latent', n_neighbors=30) sc.tl.louvain(model.adata_st, resolution=0.5) model.adata_st.obs["louvain"].to_csv(os.path.join(save_path, "clustering_result.csv")) for i, adata_st_i in enumerate(result): adata_st_i.obs["louvain"] = model.adata_st.obs.loc[adata_st_i.obs_names, ]["louvain"] color = ['#7570b3', '#1b9e77', '#d95f02', '#66a61e', '#e7298a'] plt.rcParams["figure.figsize"] = (4, 4) for i, adata_st_i in enumerate(result): sc.pl.spatial(result[i], img_key="hires", color=["louvain"], palette=color, size=1.5)
in this part "Run STitch3D on the human heart dataset". The following pictures are my encountered problems:Thanks in advance!