Add global decoder ablation workflow by sacpis · Pull Request #17 · NVIDIA/Ising-Decoding

sacpis · 2026-03-18T21:26:45Z

Add global decoder ablation workflow

Adding code/evaluation/failure_analysis.py from the internal repo
- Made a few tweaks
Adding a workflow task decoder_ablation entry point
Fixing a bug where meas_basis_test: both in the config caused an AttributeError
- Overriding the datapipe config copy with the single basis

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

…ion, and plotting helpers Signed-off-by: Sachin Pisal <spisal@nvidia.com>

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

bmhowe23

Overall, this LGTM. Just one minor comment.

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

…um-predecoder into add_global_decoders

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis · 2026-03-20T03:15:48Z

Seems like unit tests are failing because of missing ldpc module.

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis · 2026-03-20T03:30:21Z

Thank you @bmhowe23 and @ivanbasov for reviewing this PR. Please let me know if we need to add/update anything else.

bmhowe23 · 2026-03-20T04:01:01Z

Seems like unit tests are failing because of missing ldpc module.

@ivanbasov does a CI script perhaps need to be updated to install the new dependency?

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

ivanbasov · 2026-03-21T01:12:09Z

Seems like unit tests are failing because of missing ldpc module.

@ivanbasov does a CI script perhaps need to be updated to install the new dependency?

sure. @sacpis , could you please update CI scripts. CI jobs fail.

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis · 2026-03-21T01:45:41Z

Seems like unit tests are failing because of missing ldpc module.

@ivanbasov does a CI script perhaps need to be updated to install the new dependency?

sure. @sacpis , could you please update CI scripts. CI jobs fail.

Done with eaed4cc.

sacpis · 2026-03-23T15:43:29Z

@bmhowe23 @ivanbasov Good to go?

* Replace proprietary license headers with Apache-2.0 Update all SPDX headers from LicenseRef-NvidiaProprietary to Apache-2.0 across all 70 tracked source files. Also updates spdx_headers.py to generate Apache-2.0 headers and replace old proprietary headers in-place. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(headers): apply Apache-2.0 headers to files added after branch cut Files added by PRs #13, #14, and #17 still carried the proprietary LicenseRef-NvidiaProprietary header. Replace with Apache-2.0 to match the rest of the codebase after the header migration. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * style: apply YAPF formatting after header replacement Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(headers): restore full file content truncated during rebase The first rebase used --theirs to resolve header conflicts, which took the old PR branch content instead of main's newer content for 5 files. Restore from upstream/main and apply Apache-2.0 header correctly. Affected files: - code/qec/noise_model.py - code/qec/surface_code/homological_equivalence_torch.py - code/tests/mid/test_homological_equivalence.py - code/tests/test_noise_model.py - code/training/train.py Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(test): remove PreDecoderModelMemory_v2 test removed by PR #18 PR #18 removed the unused v2 model architecture. Drop the corresponding test class and import to fix the ImportError in CI. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

* adding dependencies for decoder_ablation workflow Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding failure_analysis containing the decoder helpers, decoder ablation, and plotting helpers Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding modified code/evaluation/failure_analysis.py Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding tests for failure analysis Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding decoder_ablation as a workflow task Signed-off-by: Sachin Pisal <spisal@nvidia.com> * overriding the config copy with the resolved single basis Signed-off-by: Sachin Pisal <spisal@nvidia.com> * formatting Signed-off-by: Sachin Pisal <spisal@nvidia.com> * formatting Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding CUDA-Q nv-qldpc-decoder from internal repo and tests Signed-off-by: Sachin Pisal <spisal@nvidia.com> * removing _PYMATCHING_SUPPORTS_CORRELATIONS and inspect Signed-off-by: Sachin Pisal <spisal@nvidia.com> * formatting Signed-off-by: Sachin Pisal <spisal@nvidia.com> * removing unconditional imports Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding a test to check predecoder actually modifies residual syndromes Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding comments for _decode_ldpc_batch Signed-off-by: Sachin Pisal <spisal@nvidia.com> * refactoring decoder_ablation_study Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding unittests for refactored functions Signed-off-by: Sachin Pisal <spisal@nvidia.com> * tracking unavailable decoders Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding BP variants in try/except block Signed-off-by: Sachin Pisal <spisal@nvidia.com> * removing redundant check Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding modules to install to requirements Signed-off-by: Sachin Pisal <spisal@nvidia.com> --------- Signed-off-by: Sachin Pisal <spisal@nvidia.com>

* Replace proprietary license headers with Apache-2.0 Update all SPDX headers from LicenseRef-NvidiaProprietary to Apache-2.0 across all 70 tracked source files. Also updates spdx_headers.py to generate Apache-2.0 headers and replace old proprietary headers in-place. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(headers): apply Apache-2.0 headers to files added after branch cut Files added by PRs #13, #14, and #17 still carried the proprietary LicenseRef-NvidiaProprietary header. Replace with Apache-2.0 to match the rest of the codebase after the header migration. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * style: apply YAPF formatting after header replacement Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(headers): restore full file content truncated during rebase The first rebase used --theirs to resolve header conflicts, which took the old PR branch content instead of main's newer content for 5 files. Restore from upstream/main and apply Apache-2.0 header correctly. Affected files: - code/qec/noise_model.py - code/qec/surface_code/homological_equivalence_torch.py - code/tests/mid/test_homological_equivalence.py - code/tests/test_noise_model.py - code/training/train.py Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(test): remove PreDecoderModelMemory_v2 test removed by PR #18 PR #18 removed the unused v2 model architecture. Drop the corresponding test class and import to fix the ImportError in CI. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

* adding dependencies for decoder_ablation workflow Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding failure_analysis containing the decoder helpers, decoder ablation, and plotting helpers Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding modified code/evaluation/failure_analysis.py Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding tests for failure analysis Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding decoder_ablation as a workflow task Signed-off-by: Sachin Pisal <spisal@nvidia.com> * overriding the config copy with the resolved single basis Signed-off-by: Sachin Pisal <spisal@nvidia.com> * formatting Signed-off-by: Sachin Pisal <spisal@nvidia.com> * formatting Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding CUDA-Q nv-qldpc-decoder from internal repo and tests Signed-off-by: Sachin Pisal <spisal@nvidia.com> * removing _PYMATCHING_SUPPORTS_CORRELATIONS and inspect Signed-off-by: Sachin Pisal <spisal@nvidia.com> * formatting Signed-off-by: Sachin Pisal <spisal@nvidia.com> * removing unconditional imports Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding a test to check predecoder actually modifies residual syndromes Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding comments for _decode_ldpc_batch Signed-off-by: Sachin Pisal <spisal@nvidia.com> * refactoring decoder_ablation_study Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding unittests for refactored functions Signed-off-by: Sachin Pisal <spisal@nvidia.com> * tracking unavailable decoders Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding BP variants in try/except block Signed-off-by: Sachin Pisal <spisal@nvidia.com> * removing redundant check Signed-off-by: Sachin Pisal <spisal@nvidia.com> * adding modules to install to requirements Signed-off-by: Sachin Pisal <spisal@nvidia.com> --------- Signed-off-by: Sachin Pisal <spisal@nvidia.com>

* Replace proprietary license headers with Apache-2.0 Update all SPDX headers from LicenseRef-NvidiaProprietary to Apache-2.0 across all 70 tracked source files. Also updates spdx_headers.py to generate Apache-2.0 headers and replace old proprietary headers in-place. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(headers): apply Apache-2.0 headers to files added after branch cut Files added by PRs #13, #14, and #17 still carried the proprietary LicenseRef-NvidiaProprietary header. Replace with Apache-2.0 to match the rest of the codebase after the header migration. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * style: apply YAPF formatting after header replacement Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(headers): restore full file content truncated during rebase The first rebase used --theirs to resolve header conflicts, which took the old PR branch content instead of main's newer content for 5 files. Restore from upstream/main and apply Apache-2.0 header correctly. Affected files: - code/qec/noise_model.py - code/qec/surface_code/homological_equivalence_torch.py - code/tests/mid/test_homological_equivalence.py - code/tests/test_noise_model.py - code/training/train.py Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(test): remove PreDecoderModelMemory_v2 test removed by PR #18 PR #18 removed the unused v2 model architecture. Drop the corresponding test class and import to fix the ImportError in CI. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

sacpis added 7 commits March 18, 2026 11:20

adding dependencies for decoder_ablation workflow

f87a7a0

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

adding failure_analysis containing the decoder helpers, decoder ablat…

ebe77de

…ion, and plotting helpers Signed-off-by: Sachin Pisal <spisal@nvidia.com>

adding modified code/evaluation/failure_analysis.py

01b9fc0

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

adding tests for failure analysis

0a85a62

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

adding decoder_ablation as a workflow task

fe87b09

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

overriding the config copy with the resolved single basis

a7f9881

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

formatting

9519f1b

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis requested a review from bmhowe23 March 18, 2026 21:26

formatting

834da19

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

bmhowe23 reviewed Mar 18, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py

adding CUDA-Q nv-qldpc-decoder from internal repo and tests

a32b0c3

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis marked this pull request as ready for review March 18, 2026 23:40

sacpis requested a review from bmhowe23 March 18, 2026 23:53

bmhowe23 reviewed Mar 19, 2026

View reviewed changes

Comment thread code/requirements_public_train.txt

bmhowe23 reviewed Mar 19, 2026

View reviewed changes

Comment thread code/evaluation/logical_error_rate.py Outdated

bmhowe23 reviewed Mar 19, 2026

View reviewed changes

sacpis added 2 commits March 19, 2026 14:30

removing _PYMATCHING_SUPPORTS_CORRELATIONS and inspect

604e539

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

formatting

c86eb80

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis requested a review from bmhowe23 March 19, 2026 21:44

Merge branch 'main' into add_global_decoders

8c0c94a

ivanbasov reviewed Mar 19, 2026

View reviewed changes

Comment thread code/tests/test_failure_analysis.py Outdated

sacpis added 2 commits March 19, 2026 16:09

removing unconditional imports

cbb581b

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

Merge branch 'add_global_decoders' of https://github.com/sacpis/quant…

6e2fd4b

…um-predecoder into add_global_decoders

sacpis requested a review from ivanbasov March 19, 2026 23:10

ivanbasov reviewed Mar 19, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py

ivanbasov reviewed Mar 19, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py

ivanbasov reviewed Mar 19, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py

ivanbasov reviewed Mar 19, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py

ivanbasov reviewed Mar 19, 2026

View reviewed changes

Comment thread code/tests/test_failure_analysis.py

ivanbasov requested changes Mar 19, 2026

View reviewed changes

sacpis added 4 commits March 19, 2026 16:52

adding a test to check predecoder actually modifies residual syndromes

8264cbb

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

adding comments for _decode_ldpc_batch

2883c48

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

refactoring decoder_ablation_study

a261d4d

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

adding unittests for refactored functions

be74a67

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

tracking unavailable decoders

a6a52da

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis requested a review from ivanbasov March 20, 2026 03:29

ivanbasov reviewed Mar 20, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py Outdated

ivanbasov reviewed Mar 20, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py

ivanbasov reviewed Mar 20, 2026

View reviewed changes

Comment thread code/evaluation/failure_analysis.py Outdated

sacpis added 2 commits March 20, 2026 14:18

adding BP variants in try/except block

443dfa8

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

removing redundant check

9837143

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

sacpis requested a review from ivanbasov March 20, 2026 21:36

adding modules to install to requirements

eaed4cc

Signed-off-by: Sachin Pisal <spisal@nvidia.com>

ivanbasov approved these changes Mar 23, 2026

View reviewed changes

sacpis merged commit 9f6d854 into NVIDIA:main Mar 23, 2026
12 checks passed

ivanbasov mentioned this pull request Apr 1, 2026

feat(decoder_ablation): TRT pre-decoder backend + cudaq-qec docs and tests #34

Merged

3 tasks

Conversation

sacpis commented Mar 18, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

bmhowe23 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

sacpis commented Mar 20, 2026

Uh oh!

sacpis commented Mar 20, 2026

Uh oh!

bmhowe23 commented Mar 20, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

ivanbasov commented Mar 21, 2026

Uh oh!

sacpis commented Mar 21, 2026

Uh oh!

sacpis commented Mar 23, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants