Repository navigation
Tutorials/Sim2Real Blogs: Add validated box benchmarks and viewer fixes. - #283
Open
johnnynunez wants to merge 3 commits into
Open
johnnynunez wants to merge 3 commits into
johnnynunez wants to merge 3 commits into
Conversation
Refresh the MuJoCo/MJWarp lessons with the physically validated two-cube receiving-box task, matched CPU/GPU benchmark, finite viewer lifecycle, reproducible environments and unchanged raw evidence. Follow-up to NVIDIA#279. Signed-off-by: Johnny <johnnync13@gmail.com>
Bind the published notebooks and links to verified public Hub commits while retaining the measured runtime and original evidence. Signed-off-by: Johnny <johnnync13@gmail.com>
Bind the published notebooks and links to verified public Hub commits while retaining the measured runtime and original evidence. Signed-off-by: Johnny <johnnync13@gmail.com>
Contributor
|
👋 Thank you for your contribution! This pull request is from a forked repository so GitHub Actions will not be able to run CI. A maintainer will review your changes shortly and manually trigger the CI. @maintainers Please review this PR when you have a chance and follow the instructions in the CONTRIBUTING.md file to trigger the CI. |
This branch has not been deployed
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The MuJoCo/MJWarp tutorial now uses the same two-cube receiving-box task on CPU and GPU: grasp, lift, carry, release, settle and withdraw. The lessons run reference solutions by default, preserve the learner exercises, and include the viewer completion and clean-exit fixes. This follows merged PR #279.
The new benchmark notebook starts in preflight mode. Its reproducible sweep compares 1, 16, 32, 64, 128, 256, 512, 1024 and 2048 worlds for SO-101 and reBot, with one full warm-up and five measured episodes. The retained workstation study accepted 35/36 configurations; reBot GPU512 remains excluded, with the original failure and timings preserved as diagnostic evidence. Raw archives, figures, package locks, hardware/configuration records and the separate ALOHA reference are included unchanged.
The tutorial and container use the measured MuJoCo 3.8.0 / MuJoCo Warp 3.8.0.3 / Warp 1.15.0 environment. It is intentionally separate from the Newton tutorial's 3.12 stack. Scoped LFS rules preserve the evidence payloads and figures; source MIT and model notices remain.
Validation: updated Docker image built; 151 tests passed, 1 optional viewer-integration test skipped, including both CPU/GPU robot tasks and real-kernel benchmark preflight. Four notebook format checks and source/LFS hash audits pass. The initial wrong-scope asset test and its corrected rerun are retained in
hub-validation/2026-10-10.These are community measurements on documented hardware, not official product benchmarks. The new container checks validate the Hub relocation; they do not replace the retained performance sweep. Live Brev and Colab execution remain unverified.
Fresh validation record: https://github.com/johnnynunez/accelerated-computing-hub/blob/e923560f71080d807b6d3be17c069dc2694fdbbc/tutorials/sim2real-blogs/hub-validation/2026-10-10/README.md