admin

admin synced commits to refs/heads/feat/smolvla-on-steerable at admin/lerobot from mirror 2026-05-13 23:59:43 +00:00

95033733fc deps: add sentencepiece to the pi extra (FAST action tokenizer)

admin synced commits to refs/heads/docs/quickstart-fix at admin/lerobot from mirror 2026-05-13 23:59:43 +00:00

admin synced new reference refs/heads/docs/quickstart-fix to admin/lerobot from mirror 2026-05-13 23:59:43 +00:00

admin synced commits to refs/heads/docs/cheat-sheet at admin/lerobot from mirror 2026-05-13 23:59:43 +00:00

3bd1f19e71 update

admin synced commits to refs/heads/feat/codec-parameters at admin/lerobot from mirror 2026-05-13 15:49:54 +00:00

d8c17a20a0 typos(typos): typos

cf6f92ac74 fix(re-encoding): removing inconsistent re-encoding option in lerobot_edit_dataset

f0830972f0 docs(warning): adding warning regarding encoding parameters modification

d1917a4871 chore(format): formatting code

f89da582ae docs(dataset): addind details about datasets

Compare 8 commits »

admin synced commits to refs/heads/feat/vla-jepa at admin/lerobot from mirror 2026-05-13 15:49:54 +00:00

ddaff399b5 adding more tests to ensure good coverage

e28d34a3cf some more fixes to be closer to the original implem

bd90da2ce1 adjusting obs steps, tublets size to match original implementation

bb1b1ff041 fixing wm_loss not propagating

25430cf75a fix warnings with qwen processor kwargs

Compare 6 commits »

admin synced commits to refs/heads/feat/smolvla-on-steerable at admin/lerobot from mirror 2026-05-13 15:49:54 +00:00

c3503b774f fix(debug): dumper now shows real stream + target flags

99ebee4d16 annotate: tighter subtask + memory prompts (≤5 / ≤10 words)

a8ca5128b8 fix(annotate): re-emit plan at every subtask boundary

dd97c33814 refactor(annotate): plan = summary of still-todo subtasks, drop VLM call

fa45ba631b fix(policies,recipe): register PI052Config + allow flow-only sub-recipes

Compare 27 commits »

admin synced commits to refs/heads/main at admin/lerobot from mirror 2026-05-13 15:49:53 +00:00

9db9c35cb4 fix(config): add lora_alpha to PeftConfig (#3573)

fe96b28c74 Fix policy.path not working in YAML config files (#3145)

Compare 2 commits »

admin synced commits to refs/heads/auto/update-uv-lock at admin/lerobot from mirror 2026-05-13 07:39:53 +00:00

2ba0640463 chore(dependencies): update uv.lock

2438df1307 chore(dependencies): update uv.lock (#3561)

f218d5ab30 feat(episodes): adding support for metadata based episodes filtering (#3530)

04125492e4 fix(datasets): expand torchcodec platform coverage + rewrite pyav fallback for torchvision >0.26 (#3588)

e963e5a0c4 RL stack refactoring (#3075)

Compare 5 commits »

admin synced commits to refs/heads/feat/codec-parameters at admin/lerobot from mirror 2026-05-12 23:29:53 +00:00

58dedba06e typo(typo): typo

ba25a4f8db feat(codec options): passing codec options to lerobot_edit_dataset episode deletion tool

5b976cc926 fix(typos): fixing multiple typos

6c1b04b6c9 fix(deletion): reverting unwanted deletion

1846c7e847 docs(typos): fixing typos

Compare 56 commits »

admin synced commits to refs/heads/main at admin/lerobot from mirror 2026-05-12 23:29:53 +00:00

2438df1307 chore(dependencies): update uv.lock (#3561)

f218d5ab30 feat(episodes): adding support for metadata based episodes filtering (#3530)

Compare 2 commits »

admin synced commits to refs/heads/feat/smolvla-on-steerable at admin/lerobot from mirror 2026-05-12 23:29:53 +00:00

3a20ea337e feat(smolvla2-runtime): --text_min_new_tokens / --text_temperature CLI debug knobs

b6fb536460 chore(training): bump plan/memory dropout to 0.50 to force vision-grounding

bfd3bb1791 fix(smolvla2): handle batched sample indices in chat tokenizer

4908433f9a chore(training): align smolvla2_hirobot.slurm with what's actually run

6ce1f36002 fix(smolvla2): supervise high-level head with *current* subtask at every frame

Compare 21 commits »

admin synced and deleted reference refs/heads/feat/episodes-filtering at admin/lerobot from mirror 2026-05-12 23:29:53 +00:00

admin synced and deleted reference refs/heads/auto/update-uv-lock at admin/lerobot from mirror 2026-05-12 23:29:53 +00:00

admin synced commits to refs/heads/feat/smolvla-on-steerable at admin/lerobot from mirror 2026-05-12 15:19:43 +00:00

d866c2c9fd fix(smolvla2): only regenerate chunk when queue is fully drained

01e2228b24 feat(smolvla2): per-component prompt dropout + augmented training script

c36de3a3e8 fix(smolvla2): enqueue full chunk via predict_action_chunk

cbfaf2c544 feat(smolvla2): action-dispatch counter + tighter gibberish filter

d0278ea093 feat(smolvla2): render state panel in autonomous mode too

Compare 14 commits »

admin synced commits to refs/heads/docs/cheat-sheet at admin/lerobot from mirror 2026-05-12 15:19:43 +00:00

admin synced new reference refs/heads/docs/cheat-sheet to admin/lerobot from mirror 2026-05-12 15:19:43 +00:00

admin synced commits to refs/heads/main at admin/lerobot from mirror 2026-05-12 15:19:43 +00:00

04125492e4 fix(datasets): expand torchcodec platform coverage + rewrite pyav fallback for torchvision >0.26 (#3588)

e963e5a0c4 RL stack refactoring (#3075)

Compare 2 commits »

admin synced commits to refs/heads/feat/vla-jepa at admin/lerobot from mirror 2026-05-12 15:19:43 +00:00

96a8efddf0 adding guards to avoid needing transformers and diffusers for type checking and basic tests

ff7dc0519e updating uv lock

c935628316 adding deps to pyproject.toml

2e9ba42e1b linting

3144029814 (feat)policies: add VLA-JEPA

Compare 10 commits »

admin synced and deleted reference refs/heads/user/khalil-meftah/2026-02-16-rl-stack-refactor at admin/lerobot from mirror 2026-05-12 15:19:43 +00:00