ydy0615
0 Followers · 0 Following
  • Joined on 2025-12-26
Repositories
5
 Projects Packages Public Activity Starred Repositories
ydy0615 synced commits to feat/language-annotation-pipeline at ydy0615/lerobot-clone from mirror 2026-05-26 08:21:24 +00:00
1e7c0d6aa1 annotate(plan): force composite-action subtasks; ban ultra-fine splits
920c6ef5a2 docs(annotate): disable phase-0 vocabulary discovery by default in run_hf_job
Compare 2 commits »
ydy0615 synced commits to auto/update-uv-lock at ydy0615/lerobot-clone from mirror 2026-05-26 08:21:24 +00:00
31ae19ccbf chore(dependencies): update uv.lock
ydy0615 synced commits to feat/smolvla-on-steerable at ydy0615/lerobot-clone from mirror 2026-05-26 00:11:23 +00:00
4913356564 pi052: SDPA attention port + selective AC + bench harness
673cc6b0fe pi052: opt-in Liger fused kernels (rope + geglu + layer_norm)
2ed6519a93 ema: enable by default (matches openpi JAX behavior)
72ea531017 train: switch EMA from custom ModelEMA to ema-pytorch
56a934ec55 train: EMA of policy parameters (opt-in via --ema.enable=true)
Compare 7 commits »
ydy0615 synced commits to feat/smolvla-on-steerable at ydy0615/lerobot-clone from mirror 2026-05-25 16:01:25 +00:00
057c794ffe wandb: flip training-example logging defaults to on (every 5000 steps)
b1e83f556c train: periodic wandb log of training examples (images + text + actions)
da3e87ee86 Merge branch 'feat/smolvla-on-steerable' of https://github.com/huggingface/lerobot into feat/smolvla-on-steerable
1e9a6d044d Merge remote-tracking branch 'origin/feat/language-annotation-pipeline' into feat/smolvla-on-steerable
3fdfcb912a examples(port_datasets): generalize RoboCasa builder + add smoke script
Compare 115 commits »
ydy0615 synced commits to feat/language-annotation-pipeline at ydy0615/lerobot-clone from mirror 2026-05-25 16:01:25 +00:00
c37b1fc7d0 Merge origin/feat/language-annotation-pipeline (8 fix(annotate) commits + vocabulary phase)
9020635b14 Merge branch 'main' into feat/language-annotation-pipeline
8194897994 fix(deps): cap placo below 0.9.16 and harden kinematics import (#3647)
9f437d86b6 fix(groot): align GR00TN15Config with transformers config dataclasses (#3606)
b74a551d38 fix(pi0, pi05): stabilize torch.compile and expand test coverage (#3610)
Compare 34 commits »
ydy0615 synced commits to feat/add-topreward-rm at ydy0615/lerobot-clone from mirror 2026-05-25 16:01:25 +00:00
3b5b94dbd6 optmize topreward input processing (#3660)
ydy0615 synced commits to auto/update-uv-lock at ydy0615/lerobot-clone from mirror 2026-05-25 07:51:25 +00:00
b14587bada chore(dependencies): update uv.lock
ydy0615 pushed to main at ydy0615/llm-in-text 2026-05-24 15:32:09 +00:00
59334e4057 Stabilize pro editing without heavy office runtime
ydy0615 synced new reference rebot-mit-mode to ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
ydy0615 synced commits to rebot-mit-mode at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
ydy0615 synced commits to main at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
8194897994 fix(deps): cap placo below 0.9.16 and harden kinematics import (#3647)
9f437d86b6 fix(groot): align GR00TN15Config with transformers config dataclasses (#3606)
b74a551d38 fix(pi0, pi05): stabilize torch.compile and expand test coverage (#3610)
c0a2e9814d fix examples (#3623)
bac4f61eae refactor: support custom progress parquet overlays (#3640)
Compare 11 commits »
ydy0615 synced commits to feat/vla-jepa at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
b75b3ce02d fix qwen norm layer output libero eval is now as expected
5495c10cdf trying to close success rate gap
1bcba9dec6 fixing training and exal examples
dd13eda002 adding guard for diffusers
c01a00a972 adressing dtype zeros issue
Compare 52 commits »
ydy0615 synced commits to feat/smolvla-on-steerable at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
77a16db529 fix(smolvla2): make HighLevelSubtaskFwd actually fire at low hz + quiet startup log
ca1b951e7b feat(pi05): expose lm_head_lr_scale for stronger text-CE gradient
9d30d91021 fix(pi052,smolvla2): unblock text generation when LM head drifted to <loc>
e050d0fe0a fix(recipes): use active_at for memory_update, rebalance subtask_mem
2ca030fa28 fix(pi052): build processors from current config
Compare 27 commits »
ydy0615 synced commits to feat/language-annotation-pipeline at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
471b2b1b1d fix(annotate): bump same-frame subtasks onto distinct frames
a15e16c072 fix(annotate): replace fuzzy subtask snapping with strict match + one-shot retry
336af85c09 fix(annotate): never leave an episode with zero canonical subtasks
54221ceea2 feat(annotate): let the VLM decide vocabulary size
369ab17110 fix(annotate): update run_hf_job CLI args for renamed namespaces + phase 0
Compare 11 commits »
ydy0615 synced new reference feat/depth-integration to ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
ydy0615 synced commits to feat/depth-integration at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
ydy0615 synced new reference feat/add-topreward-rm to ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
ydy0615 synced commits to feat/add-topreward-rm at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
ydy0615 synced commits to feat/add-robometer-rm at ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00
a640d752bf chore(rewards/robometer): default to lerobot/Robometer-4b model
1b559e1c84 fix(rewards/robometer): add missing input keys mm_token_type_ids
bcd1ace38d feat(rewards): add Robometer offline progress labeling script
Compare 3 commits »
ydy0615 synced new reference docs/model-card-improvements to ydy0615/lerobot-clone from mirror 2026-05-24 15:31:30 +00:00