Add color-code pre-decoder support#89
Merged
Merged
Conversation
ivanbasov
force-pushed
the
color-code-support
branch
4 times, most recently
from
8180190 to
c730455
Compare
Add color-code support to the AI pre-decoder stack: triangular color-code memory circuits with augmented-DEM precompute and homological-equivalence canonicalization, Chromobius as the global decoder, a unified surface/color runner, training and threshold configs, docs and cookbook tutorial, and color-code unit tests. The weight-6 homological-equivalence tests are aligned with the canonical orientation actually implemented by WEIGHT6_WEIGHT3_RULES, which is frozen for compatibility with trained checkpoints and published results. Tests that compare against the legacy reference circuit generator skip when that reference is not available. No pre-trained color-code checkpoint is included in this release; inference and threshold sweeps run on user-trained Model-5-shaped checkpoints. Co-authored-by: Jan Olle <jolleaguiler@nvidia.com> Co-authored-by: Kevin Mato <kmato@nvidia.com> Co-authored-by: Chris Chamberland <cchamberland@nvidia.com> Co-authored-by: Muyuan Li <muyuanl@nvidia.com> Co-authored-by: Igor Baratta <ialmeidabara@nvidia.com>
ivanbasov
force-pushed
the
color-code-support
branch
from
c730455 to
0201164
Compare
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This release adds color-code support to the AI pre-decoder stack:
code/qec/color_code/): triangular color-code memorycircuits (superdense, nearest-neighbor schedule), augmented-DEM precompute
(
code/qec/precompute_dem.py --code color), homological-equivalencecanonicalization, and Chromobius as the global decoder on a Torch +
cuStabilizer runtime.
code: surface | colorinconf/config_public.yaml,same
local_run.sh/workflows/run.pyentry points; color dispatches tothe same
inference/threshold/sdr/chromobius_timingtasks.conf/config_color_model_1_s_LR3e-4.yaml(training),conf/config_color_threshold_model_1_d13.yaml(threshold sweep),conf/config_inference_color_model_5.yaml(standalone inference; setmodel_checkpoint_fileto your trained checkpoint).cookbook/color_code.ipynbtutorial, results and latency figures.code/tests/, public inference testtest_inference_public_color_model.py, test-tier script updates. Theweight-6 homological-equivalence tests assert the canonical orientation
actually implemented by
WEIGHT6_WEIGHT3_RULES, which is frozen forcompatibility with trained checkpoints and published results. Tests that
compare against the legacy reference circuit generator skip when that
reference is not available.
No pre-trained color-code checkpoint is included in this release; inference
and threshold sweeps run on user-trained Model-5-shaped checkpoints (see
README → Color code support).