Commit Graph

  • 0eb9b5d1a5 Sanitize cfg.wandb Cadene 2024-02-25 11:15:09 +0000
  • 5812298290 fix style readme Cadene 2024-02-25 11:09:47 +0000
  • e765e26b0b Sanitize cfg.policy, Fix skip_frame pusht.yaml Cadene 2024-02-25 11:09:02 +0000
  • fc4b98544b Add tests to Readme Cadene 2024-02-25 10:52:31 +0000
  • 6f5c731936 Rename test -> tests Cadene 2024-02-25 10:51:07 +0000
  • 598bb496b0 Add policies/factory, Add test, Add _self_ in config Cadene 2024-02-25 10:50:23 +0000
  • 64b5920e94 format Cadene 2024-02-24 18:19:18 +0000
  • aed02dc7c6 Add multithreading for video generation, Speed policy sampling Cadene 2024-02-24 18:18:39 +0000
  • 591985c67d Fix done in pusht, Fix --time in sbatch Cadene 2024-02-22 17:58:26 +0000
  • 664cfb2023 Add sbatch.sh Cadene 2024-02-22 13:04:32 +0000
  • 63d18475cc fix simxarm factory Cadene 2024-02-22 13:04:24 +0000
  • 96c53ad06f remove comments Cadene 2024-02-22 12:15:14 +0000
  • e3643d6146 Wandb works, One output dir Cadene 2024-02-22 12:14:12 +0000
  • ece89730e6 Add pusht dataset (TODO verify reward is aligned), Refactor visualize_dataset, Add video_dir, fps, state_dim, action_dim to config (Training works) Cadene 2024-02-21 00:49:40 +0000
  • 3dc14b5576 Add Prod transform, Add test_factory Cadene 2024-02-20 14:22:16 +0000
  • 3da6ffb2cb Fix unit tests, Refactor, Add pusht env, (TODO pusht replay buffer, image preprocessing) Cadene 2024-02-20 12:26:57 +0000
  • fdfb2010fd black Cadene 2024-02-18 01:24:19 +0000
  • a5c305a7a4 offline training + online finetuning converge to 33 reward! Cadene 2024-02-18 01:23:44 +0000
  • 0b4084f0f8 Clean + alpha beta corresponds to config (before 0.7 and 0.9) Cadene 2024-02-16 16:27:54 +0000
  • 0cdd23dcac Update README Cadene 2024-02-16 15:14:59 +0000
  • c202c2b3c2 Online finetuning runs (sometimes crash because of nans) Cadene 2024-02-16 15:13:24 +0000
  • 228c045674 Eval reproduced! Train running (but not reproduced) Cadene 2024-02-10 15:46:24 +0000
  • 937b2f8cba Add option for random policy Cadene 2024-01-31 13:54:32 +0000
  • 5a5b190f70 Add common, refactor eval with eval_policy Cadene 2024-01-31 13:48:12 +0000
  • 1e52499490 eval.mp4 works! Cadene 2024-01-30 23:30:14 +0000
  • 1144819c29 First real commit, simxarm env added with torchrl! Cadene 2024-01-29 12:49:30 +0000
  • 0396980450 .gitignore Cadene 2024-01-29 12:49:06 +0000
  • 007ffa898f first commit Cadene 2024-01-26 15:51:11 +0000