Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,6 @@ configuration entry point; the selected owner is materialized into the shared
| CLI Task | Registered Env | Default Motion | Owner Evidence |
| --- | --- | --- | --- |
| `g1_motion_tracking` | `G1MotionTracking` | `dance1_subject2_part.npz` | `src/unilab/conf/ppo/task/g1_motion_tracking/`, `src/unilab/conf/appo/task/g1_motion_tracking/` |
| `g1_flip_tracking` | `G1FlipTracking` | `flip_360_001__A304.npz` | `src/unilab/conf/ppo/task/g1_flip_tracking/`, `src/unilab/conf/appo/task/g1_flip_tracking/` |
| `x2_wall_flip_tracking` | `X2WallFlipTracking` | `tictacflip_6-3_g1format.npz` | `src/unilab/conf/ppo/task/x2_wall_flip_tracking/` |
| `g1_box_tracking` | `G1BoxTracking` | `sub3_largebox_003_boxconverted.npz` | `src/unilab/conf/ppo/task/g1_box_tracking/` |
| `g1_wbt_obs` | `G1WBTObs` | `dance1_subject2_part.npz` | `src/unilab/conf/sac/task/g1_wbt_obs/mujoco.yaml` |
Expand All @@ -33,13 +32,10 @@ profiles now live in `unitree_rl_unilab`.
## PPO And APPO

PPO owner iteration budgets (the `--sim mujoco` owner YAMLs): `g1_motion_tracking`
runs `algo.max_iterations=15000`; `g1_flip_tracking` runs `20000`; and
`x2_wall_flip_tracking` runs `9500`. (The Motrix owner YAML for
`g1_flip_tracking` raises this to `30000`.)
runs `algo.max_iterations=15000` and `x2_wall_flip_tracking` runs `9500`.

```bash
uv run train --algo ppo --task g1_motion_tracking --sim mujoco
uv run train --algo ppo --task g1_flip_tracking --sim mujoco
uv run train --algo ppo --task x2_wall_flip_tracking --sim mujoco
uv run train --algo ppo --task g1_motion_tracking --sim mujoco
uv run train --algo appo --task g1_motion_tracking --sim mujoco training.no_play=true
Expand Down
3 changes: 0 additions & 3 deletions docs/sphinx/source/en/5-reference/5-support_matrix.md
Original file line number Diff line number Diff line change
Expand Up @@ -85,7 +85,6 @@ This table is derived from UniSim's SDK-free public static inventory. It describ
| PPO (torch) | `go2_joystick_flat` (Go2 joystick) | Tested | - | - | - | - |
| PPO (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered | Tested | Registered | Configured |
| PPO (torch) | `g1_motion_tracking` (G1 motion tracking) | Tested | - | - | - | - |
| PPO (torch) | `g1_flip_tracking` (G1 flip tracking) | Tested | - | - | - | - |
| PPO (torch) | `x2_wall_flip_tracking` (X2 wall flip tracking) | Tested | - | - | - | - |
| PPO (torch) | `allegro_inhand` (Allegro in-hand) | Tested | - | - | - | - |
| PPO (torch) | `allegro_inhand_grasp` (allegro inhand grasp) | Tested | - | - | - | - |
Expand All @@ -94,11 +93,9 @@ This table is derived from UniSim's SDK-free public static inventory. It describ
| APPO (torch) | `go2_joystick_flat` (Go2 joystick) | Tested | - | - | - | - |
| APPO (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered | Registered | Registered | Registered |
| APPO (torch) | `g1_motion_tracking` (G1 motion tracking) | Tested | - | - | - | - |
| APPO (torch) | `g1_flip_tracking` (G1 flip tracking) | Tested | - | - | - | - |
| APPO (torch) | `allegro_inhand` (Allegro in-hand) | Tested | - | - | - | - |
| SAC (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Tested | Tested | Tested | Tested |
| SAC (torch) | `g1_motion_tracking` (G1 motion tracking) | Tested | Registered | Configured | Registered | Configured |
| SAC (torch) | `g1_flip_tracking` (G1 flip tracking) | Tested | - | Configured | - | - |
| SAC (torch) | `g1_wbt_obs` (g1 wbt obs) | Tested | - | - | - | - |
| FlashSAC (torch) | `go2_joystick_flat` (Go2 joystick) | Tested | - | - | - | - |
| FlashSAC (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered | Configured | Registered | Registered |
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,6 @@ runtime 执行。
| CLI Task | Registered Env | 默认 motion | Owner Evidence |
| --- | --- | --- | --- |
| `g1_motion_tracking` | `G1MotionTracking` | `dance1_subject2_part.npz` | `src/unilab/conf/ppo/task/g1_motion_tracking/`, `src/unilab/conf/appo/task/g1_motion_tracking/` |
| `g1_flip_tracking` | `G1FlipTracking` | `flip_360_001__A304.npz` | `src/unilab/conf/ppo/task/g1_flip_tracking/`, `src/unilab/conf/appo/task/g1_flip_tracking/` |
| `x2_wall_flip_tracking` | `X2WallFlipTracking` | `tictacflip_6-3_g1format.npz` | `src/unilab/conf/ppo/task/x2_wall_flip_tracking/` |
| `g1_box_tracking` | `G1BoxTracking` | `sub3_largebox_003_boxconverted.npz` | `src/unilab/conf/ppo/task/g1_box_tracking/` |
| `g1_wbt_obs` | `G1WBTObs` | `dance1_subject2_part.npz` | `src/unilab/conf/sac/task/g1_wbt_obs/mujoco.yaml` |
Expand All @@ -30,13 +29,10 @@ deploy 与 23-DoF motion production profile 现在位于 `unitree_rl_unilab`。
## PPO 与 APPO

PPO owner 迭代预算(`--sim mujoco` owner YAML):`g1_motion_tracking` 为
`algo.max_iterations=15000`;`g1_flip_tracking` 为 `20000`;
`x2_wall_flip_tracking` 为 `9500`。(`g1_flip_tracking` 的 Motrix owner
YAML 将其提到 `30000`。)
`algo.max_iterations=15000`,`x2_wall_flip_tracking` 为 `9500`。

```bash
uv run train --algo ppo --task g1_motion_tracking --sim mujoco
uv run train --algo ppo --task g1_flip_tracking --sim mujoco
uv run train --algo ppo --task x2_wall_flip_tracking --sim mujoco
uv run train --algo ppo --task g1_motion_tracking --sim mujoco
uv run train --algo appo --task g1_motion_tracking --sim mujoco training.no_play=true
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -30,15 +30,13 @@ uv run scripts/audit_sim2sim_contracts.py
|---|---|---|
| allegro_inhand · allegro_inhand_grasp · g1_motion_tracking · go2_handstand · go2_joystick_flat | ✅ | 无 |
| g1_box_tracking | ❌ | `obs_normalization` false↔true;`obs_groups` critic 组差异 |
| g1_flip_tracking | ❌ | `obs_normalization` true↔false;`action_scale` 29 维↔默认 0.25;`sampling_mode` 两后端运行时同为 `start`(无害) |
| g1_walk_flat | ❌ | `env.actions.joint_pos.scale` 0.25↔0.5;`obs_normalization` false↔true;`obs_groups` |

## `src/unilab/conf/appo/task/`

| Task | 判定 | 分歧 |
|---|---|---|
| allegro_inhand · g1_motion_tracking · go2_joystick_flat | ✅ | 无 |
| g1_flip_tracking | ❌ | `action_scale` 29 维↔默认 0.25;`sampling_mode` 同为 `start`(无害) |
| g1_walk_flat | ⚪ | 仅 mujoco |

已迁移的 Unitree owner 由 `unitree_rl_unilab` 自行维护 contract audit;核心状态表不再声称覆盖它们。
Expand Down
3 changes: 0 additions & 3 deletions docs/sphinx/source/zh_CN/5-reference/5-support_matrix.md
Original file line number Diff line number Diff line change
Expand Up @@ -74,7 +74,6 @@ uv run scripts/generate_support_matrix.py --write
| PPO (torch) | `go2_joystick_flat` (Go2 joystick) | Tested | - | - | - | - |
| PPO (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered | Tested | Registered | Configured |
| PPO (torch) | `g1_motion_tracking` (G1 motion tracking) | Tested | - | - | - | - |
| PPO (torch) | `g1_flip_tracking` (G1 flip tracking) | Tested | - | - | - | - |
| PPO (torch) | `x2_wall_flip_tracking` (X2 wall flip tracking) | Tested | - | - | - | - |
| PPO (torch) | `allegro_inhand` (Allegro in-hand) | Tested | - | - | - | - |
| PPO (torch) | `allegro_inhand_grasp` (allegro inhand grasp) | Tested | - | - | - | - |
Expand All @@ -83,11 +82,9 @@ uv run scripts/generate_support_matrix.py --write
| APPO (torch) | `go2_joystick_flat` (Go2 joystick) | Tested | - | - | - | - |
| APPO (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered | Registered | Registered | Registered |
| APPO (torch) | `g1_motion_tracking` (G1 motion tracking) | Tested | - | - | - | - |
| APPO (torch) | `g1_flip_tracking` (G1 flip tracking) | Tested | - | - | - | - |
| APPO (torch) | `allegro_inhand` (Allegro in-hand) | Tested | - | - | - | - |
| SAC (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Tested | Tested | Tested | Tested |
| SAC (torch) | `g1_motion_tracking` (G1 motion tracking) | Tested | Registered | Configured | Registered | Configured |
| SAC (torch) | `g1_flip_tracking` (G1 flip tracking) | Tested | - | Configured | - | - |
| SAC (torch) | `g1_wbt_obs` (g1 wbt obs) | Tested | - | - | - | - |
| FlashSAC (torch) | `go2_joystick_flat` (Go2 joystick) | Tested | - | - | - | - |
| FlashSAC (torch) | `g1_walk_flat` (G1 walk flat) | Tested | Registered | Configured | Registered | Registered |
Expand Down
2 changes: 0 additions & 2 deletions scripts/tools/support_matrix.py
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,6 @@
"go2_joystick_flat": 0,
"g1_walk_flat": 1,
"g1_motion_tracking": 2,
"g1_flip_tracking": 3,
"x2_wall_flip_tracking": 4,
"allegro_inhand": 5,
"allegro_sac": 6,
Expand All @@ -107,7 +106,6 @@
"go2_joystick_flat": "Go2 joystick",
"g1_walk_flat": "G1 walk flat",
"g1_motion_tracking": "G1 motion tracking",
"g1_flip_tracking": "G1 flip tracking",
"x2_wall_flip_tracking": "X2 wall flip tracking",
"allegro_inhand": "Allegro in-hand",
"allegro_sac": "Allegro SAC in-hand",
Expand Down
117 changes: 0 additions & 117 deletions src/unilab/conf/appo/task/g1_flip_tracking/mujoco.yaml

This file was deleted.

110 changes: 0 additions & 110 deletions src/unilab/conf/ppo/task/g1_flip_tracking/mujoco.yaml

This file was deleted.

15 changes: 0 additions & 15 deletions src/unilab/conf/sac/task/g1_flip_tracking/mjwarp_tensor.yaml

This file was deleted.

Loading