pyucell.get_rankings

Contents

pyucell.get_rankings#

pyucell.get_rankings(data, layer=None, max_rank=1500, ties_method='average', device='cpu')#

Compute per-cell ranks of genes for an AnnData object.

Parameters:

  • data (AnnData | np.ndarray | sparse matrix) – Either an AnnData object (cells x genes) or directly a 2D matrix.

  • layer (str, optional) – Only used if input is AnnData. Which layer to use (None = adata.X).

  • max_rank (int, optional) – Cap ranks at this value (ranks > max_rank are dropped for sparsity).

  • ties_method (str, optional) – Passed to scipy.stats.rankdata on the CPU path. The torch backend (device != None) only supports "min" and "ordinal".

  • device (str | None, optional) – "cpu" or None (default): CPU path with joblib parallelism (avoids PyTorch). "cuda" / "mps" / "auto": PyTorch backend for hardware acceleration. Requires the pyucell[gpu] extra.

Return type:

csr_matrix

Returns:

ranks : csr_matrix of shape (genes, cells) Sparse matrix of ranks.

Contents