pqdata

Experimental Parquet-based I/O for scverse data structures.

Installation

pip install pqdata
# or
pip install git+https://github.com/gtca/pqdata

[!WARNING] This package is experimental, and API can change between versions as well as the file structure.

Motivation

TODO

Features and integrations

TODO

I/O

Example notebook. I/O with pqdata works like this:

from pqdata import write_anndata, write_mudata
write_anndata(adata, "pbmc3k_anndata.pqdata")
write_mudata(mdata, "pbmc5k_citeseq_mudata.pqdata")

from pqdata import read_anndata, read_mudata
adata = read_anndata("pbmc3k_anndata.pqdata")
mdata = read_mudata("pbmc5k_citeseq_mudata.pqdata")