Skip to content

perf: enable apply_sparse_operator on Delaunay JAX profilers #65

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
Jammy2211 merged 1 commit into from
May 16, 2026
Merged

perf: enable apply_sparse_operator on Delaunay JAX profilers #65

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
7 changes: 6 additions & 1 deletion jax_profiling/jit/datacube/delaunay.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -211,6 +211,11 @@ def jit_profile(func, label, *args, n_repeats=10):
)

with timer.section("dataset_list_load"):
# apply_sparse_operator: precompute the NUFFT precision-matrix preload per
# channel so per-fit curvature assembly uses the FFT-based sparse path
# instead of dense DFT for every source pixel. Unblocked by PyAutoArray#316
# (the Pmax > 1 extent-indexing fix); on Delaunay this was previously
# guarded with NotImplementedError.
dataset_list = [
al.Interferometer.from_fits(
data_path=dataset_path / "data.fits",
Expand All @@ -222,7 +227,7 @@ def jit_profile(func, label, *args, n_repeats=10):
# the JAX-traceable path is the goal, NUFFT (pynufft) is not yet
# JIT-friendly.
raise_error_dft_visibilities_limit=False,
)
).apply_sparse_operator(use_jax=True, show_progress=False)
for _ in range(n_channels)
]

Expand Down
7 changes: 7 additions & 0 deletions jax_profiling/jit/interferometer/delaunay.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -212,6 +212,13 @@ def jit_profile(func, label, *args, n_repeats=10):
raise_error_dft_visibilities_limit=False,
)

with timer.section("apply_sparse_operator"):
# Precompute the NUFFT precision-matrix preload so per-fit curvature
# assembly uses the FFT-based sparse path instead of dense DFT for every
# source pixel. Unblocked by PyAutoArray#316 (the Pmax > 1 extent-indexing
# fix); on Delaunay this was previously guarded with NotImplementedError.
dataset = dataset.apply_sparse_operator(use_jax=True, show_progress=True)

n_visibilities = dataset.uv_wavelengths.shape[0]
print(f" Total visibilities: {n_visibilities}")

Expand Down