Commit Graph

300 Commits

Author SHA1 Message Date
Alexander Soare 6d0a45a97d ready for review 2024-04-12 11:36:52 +01:00
Alexander Soare 5666ec3ec7 backup wip 2024-04-11 18:33:54 +01:00
Alexander Soare 94cc22da9e Merge remote-tracking branch 'upstream/main' into refactor_dp 2024-04-11 17:52:10 +01:00
Alexander Soare 976a197f98 backup wip 2024-04-11 17:51:35 +01:00
Cadene c1a618e567 fix pusht images type from float32 to uint8, update gym-pusht dependencies 2024-04-11 14:29:16 +00:00
Cadene 657b27cc8f fix load_data_with_delta_timestamps and add tests 2024-04-11 13:00:09 +00:00
Remi 9229226522
Update lerobot/common/envs/utils.py 2024-04-11 10:35:17 +02:00
Cadene 3914831585 remove __name__ outside script 2024-04-10 17:16:44 +00:00
Cadene f8c5a2eb10 remove comment 2024-04-10 17:14:02 +00:00
Cadene 9874652c2f enable test_compute_stats
enable test_compute_stats
2024-04-10 17:12:54 +00:00
Cadene 0f0113a7a6 print_cuda_memory_usage docstring 2024-04-10 16:03:39 +00:00
Cadene e8622154f8 Replace import gym_pusht in pusht dataset by dynamic import 2024-04-10 15:56:18 +00:00
Cadene c08003278e test_examples are passing 2024-04-10 13:45:45 +00:00
Cadene 7c8eb7ff19 Merge remote-tracking branch 'origin/user/rcadene/2024_03_31_remove_torchrl' into user/rcadene/2024_03_31_remove_torchrl 2024-04-10 11:34:51 +00:00
Cadene 06573d7f67 online training works (loss goes down), remove repeat_action, eval_policy outputs episodes data, eval_policy uses max_episodes_rendered 2024-04-10 11:34:01 +00:00
Alexander Soare 575891e8ac Merge remote-tracking branch 'upstream/user/rcadene/2024_03_31_remove_torchrl' into refactor_act 2024-04-09 15:19:29 +01:00
Simon Alibert d9019d9e7e disable env_checker in factory 2024-04-09 10:24:28 +02:00
Alexander Soare e6c6c2367f Merge remote-tracking branch 'upstream/user/rcadene/2024_03_31_remove_torchrl' into refactor_act 2024-04-09 08:36:28 +01:00
Cadene 6902e01db0 tests are passing for aloha/act policies, removes abstract policy 2024-04-09 03:28:56 +00:00
Cadene 73dfa3c8e3 tests for tdmpc and diffusion policy are passing 2024-04-09 02:50:32 +00:00
Alexander Soare 50e4c8050c Merge remote-tracking branch 'upstream/user/rcadene/2024_03_31_remove_torchrl' into refactor_act 2024-04-08 17:13:11 +01:00
Cadene 1149894e1d rename handle -> task 2024-04-08 14:54:52 +00:00
Alexander Soare 9c96349926 Merge remote-tracking branch 'upstream/user/rcadene/2024_03_31_remove_torchrl' into refactor_act 2024-04-08 15:44:00 +01:00
Simon Alibert 3f6dfa4916 Add gym-aloha, rename simxarm -> xarm, refactor 2024-04-08 16:24:11 +02:00
Simon Alibert 5dff6d8339 remove aloha 2024-04-08 16:22:13 +02:00
Cadene 70aaf1c4cb test_datasets.py are passing! 2024-04-08 14:16:57 +00:00
Alexander Soare 0b4c42f4ff typos 2024-04-08 14:59:37 +01:00
Alexander Soare 62b18a7607 Add type hints 2024-04-08 14:51:45 +01:00
Alexander Soare 86365adf9f revision 2024-04-08 14:44:46 +01:00
Alexander Soare 863f28ffd8 ready for review 2024-04-08 13:10:19 +01:00
Alexander Soare 1bab4a1dd5 Eval reproduction works with gym_aloha 2024-04-08 10:23:26 +01:00
Alexander Soare e982c732f1 Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-08 09:25:45 +01:00
Cadene e1ac5dc62f fix aloha pixels env test 2024-04-07 17:20:54 +00:00
Cadene 4371a5570d Remove latency, tdmpc policy passes tests (TODO: make it work with online RL) 2024-04-07 16:01:22 +00:00
Cadene 44656d2706 test_envs are passing 2024-04-05 23:27:12 +00:00
Alexander Soare 8d2463f45b backup wip 2024-04-05 18:46:30 +01:00
Alexander Soare ecc7dd3b17 Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-05 18:35:13 +01:00
Cadene a2d3588fca wrap dm_control aloha into gymnasium (TODO: properly seeding the env) 2024-04-05 17:17:31 +00:00
Cadene 29032fbcd3 wrap dm_control aloha into gymnasium (TODO: properly seeding the env) 2024-04-05 17:17:14 +00:00
Alexander Soare ab2286025b Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-05 18:06:00 +01:00
Alexander Soare 1e71196fe3 backup wip 2024-04-05 17:38:29 +01:00
Cadene 26602269cd test_envs.py are passing, remove simxarm and pusht directories 2024-04-05 16:21:07 +00:00
Cadene f56b1a0e16 WIP tdmpc 2024-04-05 13:40:31 +00:00
Simon Alibert ab3cd3a7ba (WIP) Add gym-xarm 2024-04-05 15:35:20 +02:00
Alexander Soare 0b8d27ff2c Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-05 12:48:11 +01:00
Cadene c17dffe944 policies/utils.py 2024-04-05 11:47:15 +00:00
Alexander Soare 8ba88ba250 Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-05 12:34:14 +01:00
Cadene a420714ee4 fix: action_is_pad was missing in compute_loss 2024-04-05 11:33:39 +00:00
Alexander Soare 4863e54ce9 Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-05 12:00:31 +01:00
Cadene ad3379a73a fix memory leak due to itertools.cycle 2024-04-05 10:59:32 +00:00
Alexander Soare 9d77f5773d Merge remote-tracking branch 'Cadene/user/rcadene/2024_03_31_remove_torchrl' into refactor_act_remove_torchrl 2024-04-05 11:41:11 +01:00
Alexander Soare edb125b351 backup wip 2024-04-05 11:03:28 +01:00
Cadene 5af00d0c1e fix train.py, stats, eval.py (training is running) 2024-04-05 09:31:39 +00:00
Alexander Soare 3a4dfa82fe backup wip 2024-04-04 18:34:41 +01:00
Cadene c93ce35d8c WIP stats (TODO: run tests on stats + cmpute them) 2024-04-04 16:36:03 +00:00
Cadene 1cdfbc8b52 WIP
WIP

WIP train.py works, loss going down

WIP eval.py

Fix

WIP (eval running, TODO: verify results reproduced)

Eval works! (testing reproducibility)

WIP

pretrained model pusht reproduces same results as torchrl

pretrained model pusht reproduces same results as torchrl

Remove AbstractPolicy, Move all queues in select_action

WIP test_datasets passed (TODO: re-enable NormalizeTransform)
2024-04-04 15:31:03 +00:00
Alexander Soare 278336a39a backup wip 2024-04-03 19:23:22 +01:00
Alexander Soare 110ac5ffa1 backup wip 2024-04-03 14:21:07 +01:00
Alexander Soare c7d70a8db9 Merge remote-tracking branch 'upstream/main' into refactor_act 2024-04-03 10:08:12 +01:00
Alexander Soare caf4ffcf65 add TODO 2024-04-03 09:56:46 +01:00
Alexander Soare c50a62dd6d clarifying math 2024-04-03 09:47:38 +01:00
Alexander Soare e9eb262293 numerically sound mean computation 2024-04-03 09:44:20 +01:00
Alexander Soare 65ef8c30d0 backup wip 2024-04-02 19:13:49 +01:00
Alexander Soare 2b928eedd4 backup wip 2024-04-02 19:11:53 +01:00
Alexander Soare a6edb85da4 Remove random sampling 2024-04-02 16:52:38 +01:00
Alexander Soare 95293d459d fix stats computation 2024-04-02 16:40:33 +01:00
Alexander Soare f1148b8c2d Merge remote-tracking branch 'upstream/main' into finish_examples 2024-04-01 11:31:31 +01:00
Simon Alibert 6bddcb647e Add test_aloha env test 2024-03-28 10:35:11 +01:00
Alexander Soare b7c9c33072 revision 2024-03-27 18:33:48 +00:00
Alexander Soare 120f0aef5c Merge remote-tracking branch 'upstream/main' into finish_examples 2024-03-27 17:52:36 +00:00
Alexander Soare 68d02c80cf Remove b/c workaround 2024-03-27 12:03:19 +00:00
Alexander Soare 011f2d27fe fix tests 2024-03-26 16:40:54 +00:00
Alexander Soare 1ed0110900 finish examples 2 and 3 2024-03-26 16:13:40 +00:00
Cadene 9ced0cf1fb unskip 2024-03-26 10:45:31 +00:00
Cadene 5a46b8a2a9 fix tests 2024-03-26 10:24:46 +00:00
Alexander Soare 1a1308d62f fix environment seeding
add fixes for reproducibility

only try to start env if it is closed

revision

fix normalization and data type

Improve README

Improve README

Tests are passing, Eval pretrained model works, Add gif

Update gif

Update gif

Update gif

Update gif

Update README

Update README

update minor

Update README.md

Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>

Update README.md

Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>

Address suggestions

Update thumbnail + stats

Update thumbnail + stats

Update README.md

Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>

Add more comments

Add test_examples.py
2024-03-26 10:10:43 +00:00
Simon Alibert c5635b7d94 Minor fixes for #47 2024-03-25 18:50:47 +01:00
Simon Alibert bcfdba109f Update pre-commit & run on all files 2024-03-25 17:29:35 +01:00
Simon Alibert 7cdd6d2450 Renamed set_seed -> set_global_seed 2024-03-25 17:19:28 +01:00
Simon Alibert 058ac991eb Add simxarm back into tests 2024-03-25 16:35:46 +01:00
Simon Alibert d3adaf1379 Add stat.pth for xarm_lift_medium 2024-03-25 15:55:45 +01:00
Simon Alibert dc89166bee Upgrade gym to gymnasium 2024-03-25 15:12:21 +01:00
Simon Alibert 5ef813ff1e Remove deprecated code 2024-03-25 13:22:49 +01:00
Simon Alibert c0833f1c2d Remove simxarm download and preproc hack 2024-03-25 12:41:17 +01:00
Simon Alibert de5c30405e fix wrong version 2024-03-25 12:35:06 +01:00
Simon Alibert 462e7469e8 Add xarm_lift_medium revision 1.0 to hub 2024-03-25 12:28:07 +01:00
Simon Alibert 127de1258d WIP 2024-03-25 12:28:07 +01:00
Cadene b905111895 fix render issue 2024-03-25 12:28:07 +01:00
Simon Alibert 0c41675986 fix __init__ import Base 2024-03-25 12:28:07 +01:00
Simon Alibert 1c24bbda3f WIP Upgrading simxam from mujoco-py to mujoco python bindings 2024-03-25 12:28:07 +01:00
Remi f3cfc8b3b4
Merge pull request #46 from huggingface/user/rcadene/2024_03_23_update_stats_v1.2
Fix bug with stats.pth + Move from cadene to lerobot + Update datasets to v1.2
2024-03-24 17:53:32 +01:00
Cadene d2ef43436c move from cadene to lerobot 2024-03-23 13:34:35 +00:00
Cadene 40f3783fca v1.2 2024-03-23 11:41:56 +00:00
Alexander Soare e698d38a35 Merge remote-tracking branch 'upstream/main' into fix_environment_seeding 2024-03-22 15:11:15 +00:00
Alexander Soare 15ff3b3af8 add fixes for reproducibility 2024-03-22 15:06:57 +00:00
Alexander Soare b9047fbdd2 fix environment seeding 2024-03-22 13:25:23 +00:00
Alexander Soare 8720c568d0 Add ability to eval hub model 2024-03-22 10:26:55 +00:00
Alexander Soare 72d3c3120b Merge remote-tracking branch 'upstream/main' into fix_pusht_diffusion 2024-03-21 10:20:52 +00:00
Alexander Soare acf1174447 ready for review 2024-03-21 10:18:50 +00:00
Simon Alibert 1bd50122be
Merge pull request #40 from huggingface/user/aliberts/2024_03_20_enable_mps_device
Enable mps backend for Apple silicon devices
2024-03-20 19:33:12 +01:00
Simon Alibert 4631d36c05 Add get_safe_torch_device in policies 2024-03-20 18:38:55 +01:00
Cadene 82e6e01651 v1.1 2024-03-20 17:34:00 +00:00
Alexander Soare d323993569 backup wip 2024-03-20 15:01:27 +00:00
Alexander Soare 4b7ec81dde remove abstracmethods, fix online training 2024-03-20 14:49:41 +00:00
Alexander Soare 32e3f71dd1 backup wip 2024-03-20 09:49:16 +00:00
Alexander Soare 5332766a82 revision 2024-03-20 09:45:45 +00:00
Alexander Soare b1ec3da035 remove internal rendering hooks 2024-03-20 09:23:23 +00:00
Alexander Soare d16f6a93b3 Merge remote-tracking branch 'upstream/main' into user/alexander-soare/multistep_policy_and_serial_env 2024-03-20 09:01:45 +00:00
Alexander Soare 4f1955edfd Clear action queue when environment is reset 2024-03-20 08:31:06 +00:00
Alexander Soare c5010fee9a fix seeding 2024-03-20 08:21:33 +00:00
Alexander Soare 18fa88475b Move reset_warning_issued flag to class attribute 2024-03-20 08:09:38 +00:00
Alexander Soare 896a11f60e backup wip 2024-03-19 18:50:04 +00:00
Cadene 7d5d99e036 Address more comments 2024-03-19 16:53:07 +00:00
Cadene 10034e85c4 Aloha done 2024-03-19 16:03:42 +00:00
Alexander Soare ea17f4ce50 backup wip 2024-03-19 16:02:09 +00:00
Cadene 6a1a29386a Add replay_buffer directory in pusht datasets + aloha (WIP) 2024-03-19 15:49:45 +00:00
Alexander Soare 88347965c2 revert dp changes, make act and tdmpc batch friendly 2024-03-18 19:18:21 +00:00
Alexander Soare bae7e7b41c Merge remote-tracking branch 'origin/main' into user/alexander-soare/multistep_policy_and_serial_env 2024-03-15 14:06:53 +00:00
Alexander Soare 3124f71ebd Merge remote-tracking branch 'origin/main' into user/alexander-soare/multistep_policy_and_serial_env 2024-03-15 14:04:23 +00:00
Alexander Soare 4ecfd17f9e fix wandb artifact name and add disable option 2024-03-15 13:56:55 +00:00
Cadene b752833f3f fix download 2024-03-15 13:19:18 +00:00
Alexander Soare a45896dc8d Merge remote-tracking branch 'origin/main' into user/alexander-soare/multistep_policy_and_serial_env 2024-03-15 13:05:35 +00:00
Cadene 5805a7ffb1 small fix in type + comments 2024-03-15 12:44:52 +00:00
Cadene 41521f7e96 self.root is Path or None + The following packages are already present in the pyproject.toml and will be skipped:
- huggingface-hub

If you want to update it to the latest compatible version, you can use `poetry update package`.
If you prefer to upgrade it to the latest available version, you can use `poetry add package@latest`.

Nothing to add.
2024-03-15 10:56:46 +00:00
Cadene b10c9507d4 Small fix 2024-03-15 00:36:55 +00:00
Cadene a311d38796 Add aloha + improve readme 2024-03-15 00:30:11 +00:00
Cadene 19730b3412 Add pusht on hf dataset (WIP) 2024-03-14 16:59:37 +00:00
Alexander Soare a222c88c99 Merge branch 'user/alexander-soare/train_pusht' into user/alexander-soare/multistep_policy_and_serial_env 2024-03-14 16:06:21 +00:00
Alexander Soare ba91976944 wip: still needs batch logic for act and tdmp 2024-03-14 15:24:10 +00:00
Alexander Soare 98484ac68e ready for review 2024-03-12 21:59:01 +00:00
Alexander Soare 9512d1d2f3 Merge branch 'main' into user/alexander-soare/train_pusht 2024-03-12 19:41:27 +00:00
Remi Cadene 9d002032d1 Add Aloha env and ACT policy
WIP Aloha env tests pass

Rendering works (fps look fast tho? TODO action bounding is too wide [-1,1])

Update README

Copy past from act repo

Remove download.py add a WIP for Simxarm

Remove download.py add a WIP for Simxarm

Add act yaml (TODO: try train.py)

Training can runs (TODO: eval)

Add tasks without end_effector that are compatible with dataset, Eval can run (TODO: training and pretrained model)

Add AbstractEnv, Refactor AlohaEnv, Add rendering_hook in env, Minor modifications, (TODO: Refactor Pusht and Simxarm)

poetry lock

fix bug in compute_stats for action normalization

fix more bugs in normalization

fix training

fix import

PushtEnv inheriates AbstractEnv, Improve factory Normalization

Add _make_env to EnvAbstract

Add call_rendering_hooks to pusht env

SimxarmEnv inherites from AbstractEnv (NOT TESTED)

Add aloha tests artifacts + update pusht stats

fix image normalization: before env was in [0,1] but dataset in [0,255], and now both in [0,255]

Small fix on simxarm

Add next to obs

Add top camera to Aloha env (TODO: make it compatible with set of cameras)

Add top camera to Aloha env (TODO: make it compatible with set of cameras)
2024-03-12 10:27:48 +00:00
Alexander Soare 87fcc536f9 wip - still need to verify full training run 2024-03-11 18:45:21 +00:00
Alexander Soare 304355c917 Merge remote-tracking branch 'origin/main' into train_pusht 2024-03-11 15:37:37 +00:00
Alexander Soare 2a01487494 early training loss as expected 2024-03-11 13:34:04 +00:00
Simon Alibert 78690d197f
Merge pull request #19 from Cadene/user/aliberts/2024_03_11_wandb_config
Configure wandb entity outside config
2024-03-11 14:17:44 +01:00
Remi fab2b3240b
Merge pull request #17 from Cadene/user/rcadene/2024_03_11_bugfix_compute_stats
Fix bugs with normalization
2024-03-11 13:44:07 +01:00
Cadene 84a1647c01 fix import 2024-03-11 12:41:14 +00:00
Cadene ccd5dc5a42 fix training 2024-03-11 12:33:15 +00:00
Simon Alibert 00fe4f4f18 Configure wandb entity outside config 2024-03-11 13:09:46 +01:00
Cadene 816b2e9d63 fix more bugs in normalization 2024-03-11 11:03:51 +00:00
Cadene a7ef4a6a33 fix bug in compute_stats for action normalization 2024-03-11 09:47:54 +00:00
Simon Alibert f54ee7cda0 Fix paths 2024-03-10 16:51:50 +01:00
Simon Alibert 134009f337 Remove init files 2024-03-10 16:38:49 +01:00
Simon Alibert 6c867d78ef Integrate pusht env from diffusion 2024-03-10 16:33:03 +01:00
Simon Alibert 302b78962c Integrate diffusion policy 2024-03-10 15:31:17 +01:00
Simon Alibert 59397fb44a Move tdmpc files 2024-03-09 18:44:36 +01:00
Simon Alibert 89eaab140b Add pusht test artifact 2024-03-09 15:36:20 +01:00
Simon Alibert f1e2837d63 fix pusht data_dir path 2024-03-08 12:26:15 +01:00
Remi Cadene 524d29aa80 fix tests 2024-03-07 13:23:22 +01:00