Repository navigation
feat: add aa.interp_2d (NumPy + JAX bilinear interpolation) - #308
Merged
Merged
Conversation
Adds autoarray/numerics/ subpackage with interp_2d.py: a 2D regular-grid bilinear interpolator with NumPy (scipy RegularGridInterpolator) and JAX (jax.scipy.ndimage.map_coordinates, order=1) backends behind a single dispatcher interp_2d(points, x_axis, y_axis, values, fill_value=0.0, xp=np). Unblocks DatasetInterp in PyAutoGalaxy for the ellipse_fitting_jax feature. Issue PyAutoArray#306. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
3 tasks
Collaborator
Author
|
Downstream PR: PyAutoLabs/PyAutoGalaxy#398 |
Relocates interp_2d from the new autoarray/numerics/ subpackage into the existing operators/ subpackage. operators/ already houses similar grid-data operations (Convolver, contour, transformer, over_sampling), so a dedicated numerics/ subpackage for a single function was over-engineering. Top-level surface is now aa.interp_2d (mirroring how aa.Convolver is exposed from operators/convolver.py). Issue PyAutoArray#306. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Jammy2211
changed the title
This was referenced Sep 26, 2026
Closed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Add a JAX-compatible 2D regular-grid bilinear interpolator at
autoarray/operators/interp_2d.py, with both NumPy (scipyRegularGridInterpolator) and JAX (jax.scipy.ndimage.map_coordinates) paths and a single dispatcher surfaced at the top level asaa.interp_2d. Step 4 of theellipse_fitting_jaxfeature —DatasetInterpin PyAutoGalaxy currently blocks JAX onscipy.interpolate, and this helper unblocks it.API Changes
Adds a new public function
autoarray.interp_2d(points, x_axis, y_axis, values, fill_value=0.0, xp=np)(also reachable asautoarray.operators.interp_2d.interp_2d). NumPy path matchesscipy.interpolate.RegularGridInterpolator(bounds_error=False, fill_value=0.0)semantics byte-for-byte; JAX path usesjax.scipy.ndimage.map_coordinates(order=1, mode='constant')and is parity-tested against the NumPy path tortol=1e-6on random in-bounds queries. No symbols removed, renamed, or changed signature. See full details below.Test Plan
python -m pytest test_autoarray/operators/test_interp_2d.py -v— 4/4 passtest_autoarray/suite continues to passFull API Changes (for automation & release notes)
Added
autoarray.interp_2d(points, x_axis, y_axis, values, fill_value=0.0, xp=np)— 2D regular-grid bilinear interpolation.pointsis(N, 2)with column 0 →x_axis, column 1 →y_axis(matching scipyRegularGridInterpolator(points=(x_axis, y_axis))convention).xp=npselects scipy;xp=jnpselectsjax.scipy.ndimage.map_coordinates. Out-of-bounds queries returnfill_value. Module:autoarray.operators.interp_2d.Removed
None.
Changed Signature
None.
Changed Behaviour
None.
Migration
N/A — purely additive.
Generated with Claude Code