Add Wavelet Operator

[schuenke]

Wavelet Operator for wavelet-reco

For wavelet transforms we use The PyTorch Wavelet Toolbox (ptwt), which extends PyWavelets (pywt). It provides GPU and gradient support via a PyTorch backend.

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
src/mrpro/algorithms/csm
inati.py	24	1	96%	44
iterative_walsh.py	15	1	93%	37
src/mrpro/algorithms/dcf
dcf_voronoi.py	53	4	92%	15, 48–49, 76
src/mrpro/algorithms/optimizers
adam.py	20	1	95%	69
src/mrpro/algorithms/reconstruction
DirectReconstruction.py	28	16	43%	51–71, 85
IterativeSENSEReconstruction.py	42	23	45%	77–78, 88–98, 113–124, 138–149
Reconstruction.py	51	24	53%	41, 53–55, 79–86, 103–114
src/mrpro/data
AcqInfo.py	128	2	98%	174, 214
CsmData.py	28	3	89%	14, 84–86
DcfData.py	44	8	82%	17, 65, 77–82
IData.py	67	9	87%	119, 125, 129, 159–167
IHeader.py	75	7	91%	75, 109, 127–131
KHeader.py	164	17	90%	24, 126–130, 157, 207, 218, 225–226, 229, 236, 275–286
KNoise.py	31	15	52%	39–52, 56–61
KTrajectory.py	69	5	93%	178–182
MoveDataMixin.py	126	14	89%	14, 109, 125, 139–141, 202, 265, 279, 358, 378–379, 396–397
QData.py	39	7	82%	42, 65–73
SpatialDimension.py	46	2	96%	64, 103
TrajectoryDescription.py	14	1	93%	23
acq_filters.py	10	1	90%	47
src/mrpro/data/_kdata
KData.py	105	16	85%	107–108, 117, 125, 179–180, 215, 220–221, 240–251
KDataRemoveOsMixin.py	29	2	93%	43, 45
KDataSelectMixin.py	20	2	90%	46, 62
KDataSplitMixin.py	48	3	94%	49, 79, 88
src/mrpro/data/traj_calculators
KTrajectoryCalculator.py	25	2	92%	23, 45
KTrajectoryIsmrmrd.py	13	2	85%	41, 50
KTrajectoryPulseq.py	29	1	97%	54
src/mrpro/operators
CartesianSamplingOp.py	50	9	82%	49–50, 55–56, 61–62, 88, 91, 114
ConstraintsOp.py	60	2	97%	46, 48
EndomorphOperator.py	51	2	96%	209, 213
FiniteDifferenceOp.py	27	2	93%	48, 113
FourierOp.py	77	1	99%	131
GridSamplingOp.py	123	9	93%	60–61, 70–71, 78–79, 82, 84, 86
LinearOperator.py	80	4	95%	32, 131, 251, 256
Operator.py	52	1	98%	21
SliceProjectionOp.py	166	8	95%	39, 46, 48, 54, 191, 212, 245, 285
WaveletOp.py	120	5	96%	152, 170, 205, 210, 233
ZeroPadOp.py	16	1	94%	30
src/mrpro/utils
Rotation.py	453	28	94%	58–66, 106, 283, 368, 370, 397, 452, 457, 460, 475, 492, 497, 640, 645, 648, 664, 668, 742, 744, 752–753, 993, 1075
filters.py	62	2	97%	44, 49
slice_profiles.py	45	6	87%	18, 34, 111–114, 147
sliding_window.py	34	1	97%	34
split_idx.py	10	2	80%	43, 47
summarize_tensorvalues.py	11	9	18%	20–29
zero_pad_or_crop.py	31	6	81%	26, 30, 54, 57, 60, 63
TOTAL	3608	287	92%

Tests	Skipped	Failures	Errors	Time
825	0 💤	0 ❌	0 🔥	1m 13s ⏱️

[ckolbPTB]

I get the following error:

worker 'gw3' crashed while running 'tests/data/test_movedatamixin.py::test_movedatamixin_convert[True-dtype1-single]'

As I did not change anything there, I was wondering if @fzimmermann89 has any ideas what this could be about?

[fzimmermann89]

Out of curiosity: pytorch wavelet toolbox changed some type hints as we suggested (on their current GitHub version)

I don't remember why we cared - does this help?

Or will this now cause issues with the current waveletop code?

[ckolbPTB]

Out of curiosity: pytorch wavelet toolbox changed some type hints as we suggested (on their current GitHub version)
I don't remember why we cared - does this help?
Or will this now cause issues with the current waveletop code?

As far as I can see this fits to your typehinting in the code so it should not be a problem.

[fzimmermann89]

some comments

[schuenke]

For testing the implementation we used some real scanner data already. Should we create an example script / notebook from our test case and directly include it in this PR?

[ckolbPTB]

Should we create an example script / notebook from our test case and directly include it in this PR?

The example we had during the hackathon was already for a FISTA + Wavelet reconstruction. I would therefore suggest to add this example to the FISTA PR.

[fzimmermann89]

@CodiumAI-Agent /review
--pr_reviewer.extra_instructions="
In the possible issues section, emphasize the following:

• Is the code logic efficient?
• Is it idiopathic pytorch code?
• Are all variable und function names clear and descriptive?
• Are additional comments required to understand the logic?
• Check the docstrings for completeness
  "
  --pr_reviewer.inline_code_comments=true
  --pr_reviewer.require_score_review=true
  --pr_reviewer.num_code_suggestions="8"

[CodiumAI-Agent]

PR Reviewer Guide 🔍

(Review updated until commit 058e475)

⏱️ Estimated effort to review: 4 🔵🔵🔵🔵⚪

🏅 Score: 85

🧪 PR contains tests

🔒 No security concerns identified

⚡ Key issues to review Efficiency Concern The implementation of the wavelet transform might not be efficient due to repeated calculations and potential redundancy in data handling, especially in the forward and adjoint methods where complex data is handled separately. Variable Clarity The variable names coeff_ptwt and coeff_mrpro in the test cases might not be immediately clear to all developers. Consider renaming these to more descriptive terms that reflect their roles in the wavelet transformation process.

[CodiumAI-Agent]

Persistent review updated to latest commit 058e475

[CodiumAI-Agent]

Consider using a dictionary or a more structured data type for managing wavelet types rather than a long list of literals. This could improve readability and maintainability. [important]