Skip to content

Support Adapt and copy for WaveletOp - #62

Merged
nHackel merged 2 commits into
JuliaImageRecon:mainfrom
shukebeta:feat/issue-103-wavelet-adapt
Nov 13, 2026
Merged

nHackel merged 2 commits into
JuliaImageRecon:mainfrom
shukebeta:feat/issue-103-wavelet-adapt

Conversation

@shukebeta

Copy link
Copy Markdown
Contributor
  • WaveletOp now returns a WaveletOpImpl that keeps the shape, the wavelet and the host arrays of the transform. Adapt.adapt and copy rebuild it for the target storage, like SamplingOp and GradientOp.
  • The wavelet transform still runs on the CPU. With GPU storage, input and output are copied through host arrays, as before; the GPU guide lists WaveletOp with this note.
  • testWaveletAdapt adapts an operator with a non-default wavelet and a singleton dimension to the test backend and back, and compares the results with the CPU operator. testWavelet runs it in Float64; the Metal job runs it in Float32.

shukebeta and others added 2 commits November 13, 2026 00:52
WaveletOp now returns a WaveletOpImpl that keeps the shape without
singleton dimensions, the wavelet and the host arrays of the transform.
Adapt.adapt and copy rebuild the operator, its closures and host arrays
from the shape and wavelet with the target storage type. The transform
still runs on the CPU; non-Vector storage copies input and output through
the host arrays, as before.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The Adapt and copy checks move into testWaveletAdapt with an element-type
argument. testWavelet calls it with Float64 as before, and test/gpu/metal.jl
loads the testOperators.jl helpers and calls it with Float32.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@codecov

codecov Bot commented Nov 12, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 84.25%. Comparing base (3f8792b) to head (a1875ef).
⚠️ Report is 5 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main      #62      +/-   ##
==========================================
+ Coverage   84.01%   84.25%   +0.24%     
==========================================
  Files          16       16              
  Lines         713      724      +11     
==========================================
+ Hits          599      610      +11     
  Misses        114      114              

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@nHackel
nHackel merged commit 2e26dfc into JuliaImageRecon:main Nov 13, 2026
6 checks passed
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.

2 participants