Myungkyu/pi0_5_rmbench_taco_wodemo_luna_b64_60k

pi05rmbenchtacowodemolunab64_60k

Low-level policy for RMBench (9 simulated tabletop tasks), fine-tuned from lerobot/pi05_base on `Myungkyu/RMBench-taco-wodemo-luna` — demonstrations with dense subtask labels from the task-specific context (offline annotation).

• —Architecture: Pi0.5 (three camera views + one keyframe slot, memory-content convention fullframev1)
• —Optimizer batch 64, 60000 steps, final checkpoint
• —Inputs: head + left/right wrist images, proprioception, the current subtask text; the keyframe slot takes a retrieved past frame when the label calls for one

Configs reference the base backbone / tokenizer by hub id or by the training site's local path — point them at your local copies before loading.