Data preparation help needed : AnnData.zarr file loading time too long #219
gauravgadhvi
started this conversation in
General
Replies: 1 comment 2 replies
-
I think it is because the matrix is being formatted as CSR (https://robbinsa.me/celldata/welchlab/HY_allMerged_AnnData_optimized.zarr/X/.zattrs). Vitessce can load CSC more efficiently than CSR sparse matrices. I just deployed a new version of the Python package (3.0.0) to PyPI (https://pypi.org/project/vitessce/#history). pip uninstall vitessce
pip install vitessce[all]==3.0.0 In this version from scipy.sparse import issparse
if issparse(adata.X):
adata.X = adata.X.tocsc() |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
I am writing this to get some help about the AnnData.zarr formatted dataset and why it wouldn't load in time on the vitessce browser. I have tried creating an AnnData.zarr file with 1000 cells and 120000 cells and in both cases the browser starts loading and never moves forward from processing stage. I can see the correct number of cells and UMAP structure loaded in the background but it never stops showing the loading circle. I don't think scalability is the issue but is there a way I can troubleshoot this? I also used the optimize_adata() function prior to dumping the anndata object to zarr.
Below is the example I am trying to test :
http://vitessce.io/#?edit=false&url=https%3A%2F%2Frobbinsa.me%2Fcelldata%2Fwelchlab%2FpyTest%2FHY_optimized_zarrConfig.json
Any help or direction is highly appreciated. Thank you!
Best,
Gaurav Gadhvi
Beta Was this translation helpful? Give feedback.
All reactions