Skip to content

perf: force over_sample_size=2 when PYAUTO_WORKSPACE_SMALL_DATASETS=1#257

Merged
Jammy2211 merged 1 commit intomainfrom
feature/small-datasets-skip-oversampling
Apr 7, 2026
Merged

perf: force over_sample_size=2 when PYAUTO_WORKSPACE_SMALL_DATASETS=1#257
Jammy2211 merged 1 commit intomainfrom
feature/small-datasets-skip-oversampling

Conversation

@Jammy2211
Copy link
Copy Markdown
Collaborator

Summary

On small test grids (15x15), adaptive oversampling with 32x32 sub-grids wastes computation. Force over_sample_size=2 in Grid2D.__init__ when PYAUTO_WORKSPACE_SMALL_DATASETS=1. Uses 2 instead of 1 to avoid Sersic singularities at pixel centres.

This is a single choke point that catches all oversampling paths — radial bins, pixelization, integer constants — both via Grid2D.apply_over_sampling() and Imaging.apply_over_sampling().

API Changes

None — internal changes only, gated behind opt-in env var.

Test Plan

  • All 722 PyAutoArray tests pass
  • imaging/simulator.py runs without errors under all smoke test env vars

🤖 Generated with Claude Code

On small test grids (15x15), adaptive oversampling with 32x32 sub-grids
wastes computation. Force over_sample_size=2 in Grid2D.__init__ when
the env var is set. Uses 2 instead of 1 to avoid Sersic singularities
at pixel centres.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@Jammy2211 Jammy2211 merged commit b93cdf0 into main Apr 7, 2026
2 checks passed
@Jammy2211 Jammy2211 deleted the feature/small-datasets-skip-oversampling branch April 7, 2026 12:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant