mirror of
https://github.com/huggingface/lerobot.git
synced 2026-07-26 11:16:00 +00:00
docs(robomme): drop prototype-branch note and move dataset to lerobot/robomme
- Remove the "Related work" block referencing the prototype branch feat/robomme-integration; the PR stands on its own. - Point all dataset references at lerobot/robomme (docs, env module docstring, RoboMMEEnvConfig docstring) — this is the canonical HF location once the dataset is mirrored. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -4,14 +4,9 @@
|
|||||||
|
|
||||||
- **16 tasks** across 4 memory-skill suites
|
- **16 tasks** across 4 memory-skill suites
|
||||||
- **1,600 training demos** (100 per task, 50 val, 50 test)
|
- **1,600 training demos** (100 per task, 50 val, 50 test)
|
||||||
- **Dataset**: [`pepijn223/robomme_data_lerobot`](https://huggingface.co/datasets/pepijn223/robomme_data_lerobot) — LeRobot v3.0, 768K frames at 10 fps
|
- **Dataset**: [`lerobot/robomme`](https://huggingface.co/datasets/lerobot/robomme) — LeRobot v3.0, 768K frames at 10 fps
|
||||||
- **Simulator**: ManiSkill / SAPIEN, Panda arm, Linux only
|
- **Simulator**: ManiSkill / SAPIEN, Panda arm, Linux only
|
||||||
|
|
||||||
> **Related work**: The initial integration of RoboMME into LeRobot was prototyped in
|
|
||||||
> [`feat/robomme-integration`](https://github.com/huggingface/lerobot/tree/feat/robomme-integration).
|
|
||||||
> This PR rebases that work onto `feat/async-vector-env`, adopts the `create_envs()` dispatch
|
|
||||||
> pattern, and adds the `LazyVectorEnv` helper for large task-count benchmarks.
|
|
||||||
|
|
||||||
## Tasks
|
## Tasks
|
||||||
|
|
||||||
| Suite | Tasks |
|
| Suite | Tasks |
|
||||||
@@ -69,12 +64,12 @@ lerobot-eval \
|
|||||||
|
|
||||||
## Dataset
|
## Dataset
|
||||||
|
|
||||||
The dataset [`pepijn223/robomme_data_lerobot`](https://huggingface.co/datasets/pepijn223/robomme_data_lerobot) is in **LeRobot v3.0 format** and can be loaded directly:
|
The dataset [`lerobot/robomme`](https://huggingface.co/datasets/lerobot/robomme) is in **LeRobot v3.0 format** and can be loaded directly:
|
||||||
|
|
||||||
```python
|
```python
|
||||||
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
|
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
|
||||||
|
|
||||||
dataset = LeRobotDataset("pepijn223/robomme_data_lerobot")
|
dataset = LeRobotDataset("lerobot/robomme")
|
||||||
```
|
```
|
||||||
|
|
||||||
### Dataset features
|
### Dataset features
|
||||||
@@ -97,7 +92,7 @@ The env wrapper exposes `front_rgb` and `wrist_rgb` as observation keys. The `fe
|
|||||||
The dataset uses `image` and `wrist_image` as column names. When fine-tuning from the dataset, you may need to rename columns to match your policy's expected input keys:
|
The dataset uses `image` and `wrist_image` as column names. When fine-tuning from the dataset, you may need to rename columns to match your policy's expected input keys:
|
||||||
|
|
||||||
```python
|
```python
|
||||||
dataset = LeRobotDataset("pepijn223/robomme_data_lerobot")
|
dataset = LeRobotDataset("lerobot/robomme")
|
||||||
# rename dataset columns if needed:
|
# rename dataset columns if needed:
|
||||||
# dataset.hf_dataset = dataset.hf_dataset.rename_column("image", "front_rgb")
|
# dataset.hf_dataset = dataset.hf_dataset.rename_column("image", "front_rgb")
|
||||||
```
|
```
|
||||||
|
|||||||
@@ -582,7 +582,7 @@ class RoboMMEEnv(EnvConfig):
|
|||||||
"""RoboMME memory-augmented manipulation benchmark (ManiSkill/SAPIEN).
|
"""RoboMME memory-augmented manipulation benchmark (ManiSkill/SAPIEN).
|
||||||
|
|
||||||
16 tasks across 4 suites: Counting, Permanence, Reference, Imitation.
|
16 tasks across 4 suites: Counting, Permanence, Reference, Imitation.
|
||||||
Dataset: pepijn223/robomme_data_lerobot (LeRobot v3.0, 1,600 episodes).
|
Dataset: lerobot/robomme (LeRobot v3.0, 1,600 episodes).
|
||||||
Benchmark: https://github.com/RoboMME/robomme_benchmark
|
Benchmark: https://github.com/RoboMME/robomme_benchmark
|
||||||
|
|
||||||
Requires: pip install 'lerobot[robomme]' (Linux only).
|
Requires: pip install 'lerobot[robomme]' (Linux only).
|
||||||
|
|||||||
@@ -9,7 +9,7 @@ RoboMME tasks:
|
|||||||
Reference: PickHighlight, VideoRepick, VideoPlaceButton, VideoPlaceOrder
|
Reference: PickHighlight, VideoRepick, VideoPlaceButton, VideoPlaceOrder
|
||||||
Imitation: MoveCube, InsertPeg, PatternLock, RouteStick
|
Imitation: MoveCube, InsertPeg, PatternLock, RouteStick
|
||||||
|
|
||||||
Dataset: pepijn223/robomme_data_lerobot (LeRobot v3.0, 1,600 episodes)
|
Dataset: lerobot/robomme (LeRobot v3.0, 1,600 episodes)
|
||||||
Install: pip install 'lerobot[robomme]' (Linux only)
|
Install: pip install 'lerobot[robomme]' (Linux only)
|
||||||
Benchmark: https://github.com/RoboMME/robomme_benchmark
|
Benchmark: https://github.com/RoboMME/robomme_benchmark
|
||||||
"""
|
"""
|
||||||
|
|||||||
Reference in New Issue
Block a user