Skip to content

Add explicit support to PCA for 'covariance_eigh' svd_solver #3296

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
flying-sheep merged 2 commits into from
Oct 18, 2024
Merged

Add explicit support to PCA for 'covariance_eigh' svd_solver #3296

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1250195
Add explicit support to PCA for `'covariance_eigh'` svd_solver
flying-sheep Oct 18, 2024
028a680
relnote
flying-sheep Oct 18, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions docs/release-notes/3296.feature.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
Add explicit support to {func}`scanpy.pp.pca` for `svd_solver='covariance_eigh'` {smaller}`P Angerer`
8 changes: 4 additions & 4 deletions src/scanpy/preprocessing/_pca/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@
SvdSolvTruncatedSVDDaskML = Literal["tsqr", "randomized"]
SvdSolvDaskML = SvdSolvPCADaskML | SvdSolvTruncatedSVDDaskML

SvdSolvPCASklearn = Literal["auto", "full", "arpack", "randomized"]
SvdSolvPCASklearn = Literal["auto", "full", "arpack", "covariance_eigh", "randomized"]
SvdSolvTruncatedSVDSklearn = Literal["arpack", "randomized"]
SvdSolvPCASparseSklearn = Literal["arpack"]
SvdSolvSkearn = SvdSolvPCASklearn | SvdSolvTruncatedSVDSklearn | SvdSolvPCASparseSklearn
Expand Down Expand Up @@ -116,13 +116,13 @@ def pca(
`'arpack'`
for the ARPACK wrapper in SciPy (:func:`~scipy.sparse.linalg.svds`)
Not available with *dask* arrays.
`'covariance_eigh'`
Classic eigendecomposition of the covariance matrix, suited for tall-and-skinny matrices.
`'randomized'`
for the randomized algorithm due to Halko (2009). For *dask* arrays,
this will use :func:`~dask.array.linalg.svd_compressed`.
`'auto'`
chooses automatically depending on the size of the problem.
`'lobpcg'`
An alternative SciPy solver. Not available with dask arrays.
`'tsqr'`
Only available with *dask* arrays. "tsqr"
algorithm from Benson et. al. (2013).
Expand All @@ -133,7 +133,7 @@ def pca(
Default value changed from `'auto'` to `'arpack'`.

Efficient computation of the principal components of a sparse matrix
currently only works with the `'arpack`' or `'lobpcg'` solvers.
currently only works with the `'arpack`' or `'covariance_eigh`' solver.

If X is a *dask* array, *dask-ml* classes :class:`~dask_ml.decomposition.PCA`,
:class:`~dask_ml.decomposition.IncrementalPCA`, or
Expand Down
4 changes: 3 additions & 1 deletion tests/test_pca.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@
from scipy.sparse import issparse

import scanpy as sc
from scanpy.preprocessing._pca import SvdSolver as SvdSolverSupported
from testing.scanpy import _helpers
from testing.scanpy._helpers.data import pbmc3k_normalized
from testing.scanpy._pytest.marks import needs
Expand Down Expand Up @@ -110,7 +111,8 @@ def array_type(request: pytest.FixtureRequest) -> ArrayType:
return request.param


SVDSolver = Literal["auto", "full", "arpack", "randomized", "tsqr", "lobpcg"]
SVDSolverDeprecated = Literal["lobpcg"]
SVDSolver = SvdSolverSupported | SVDSolverDeprecated


def gen_pca_params(
Expand Down