Aagam Shah comments

Results 14 comments of


                                            Aagam Shah

Error with `import10xFeatureMatrix` when reading 6 files: rownames (genes) of individual RNA objects are not equivalent

Hey @rcorces any idea when `1.0.3` would be released?

Question on computing niches on spatial data

Thanks for the quick response! I did run the function with `batch_key` and I computed the neighbors using `sc.pp.neighbors` and louvain clusters with `sc.tl.louvain` with `resolution=0.5` and it gave me...

Explore options to run multi GPU cugraph leiden for a large graph using some host memory to reduce GPU count

Thanks @ChuckHastings that's what I thought. But, realistically with the advances in single-cell sequencing and spatial transcriptomics the problem is going to get worse. Right now, we have a 120...

Explore options to run multi GPU cugraph leiden for a large graph using some host memory to reduce GPU count

Hey @rlratzel is it different than setting `enable_cudf_spill = True` while initializing local cuda cluster? I do that in my code above already.

[FEA]: Fix labels for a subset of nodes in Leiden clustering

Adding more clarity here. I want to provide the initial membership labels for all nodes but fix them for a subset of the nodes. Here's how it's done currently using...

UMAP on Billions of Data Points + Sample Weighting

@ghiggi if you have access to a NVIDIA GPU, you can try computing nearest neighbors with CAGRA from cuVS. It's very fast and the recall was excellent on my data...

[FEA]: Multiplex leiden clustering

~~I would also consider adding support for directed weighted graphs since `scanpy.tl.leiden` uses a directed weighted graph with `leidenalg` package.~~(Nevermind since they are moving to `igraph`). Also, support for fixing...

[BUG] umap needs init_pos = 'random' for larger datasets

@Intron7 https://github.com/rapidsai/cuml/issues/6045. Unless this issue is fixed, we should remove `init_pos=spectral` for all UMAP computations.

Updating memberships with `is_membership_fixed` is very slow

Hey, sorry about that. To be specific, I have a undirected weighted graph with 130 million nodes and 2.35 billion edges. I fixed labels for 47.1 million nodes for the...

Updating memberships with `is_membership_fixed` is very slow

@SamGG since this could be a bug that limits usage, can we prioritize fixing this?