Ville Kuosmanen eaf0218bc8 feat(policy): use pretrained vision encoder weights by default for diffusion and vqbet (#3202) ...

* feat: add pretrained vision encoder weights for diffusion and vqbet

* fix test by re-generating artifacts

---------

Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>

2026-05-07 12:10:38 +02:00

..

aloha_sim_insertion_human_act_

feat(processors): use pipelines across the codebase (#1452)

2025-09-18 15:25:26 +02:00

aloha_sim_insertion_human_act_1000_steps

fix(policy): loss normalization for padded actions in ACT, Diffusion, and MultiTaskDiT (#3442)

2026-04-23 15:23:54 +02:00

pusht_diffusion_

feat(policy): use pretrained vision encoder weights by default for diffusion and vqbet (#3202)

2026-05-07 12:10:38 +02:00

xarm_lift_medium_tdmpc_use_mpc

feat(processors): use pipelines across the codebase (#1452)

2025-09-18 15:25:26 +02:00

xarm_lift_medium_tdmpc_use_policy

feat(processors): use pipelines across the codebase (#1452)

2025-09-18 15:25:26 +02:00

save_policy_to_safetensors.py

feat(dataset): 2x faster dataloader via parallel decode, uint8 transport, and persistent workers (#3406)

2026-04-19 00:08:22 +02:00