hangX/lerobot
1
0
Fork 0
forked from tangger/lerobot
Code Pull Requests Activity

Package folder structure (#1417)

Browse Source 
* Move files

* Replace imports & paths

* Update relative paths

* Update doc symlinks

* Update instructions paths

* Fix imports

* Update grpc files

* Update more instructions

* Downgrade grpc-tools

* Update manifest

* Update more paths

* Update config paths

* Update CI paths

* Update bandit exclusions

* Remove walkthrough section
This commit is contained in:
Simon Alibert
2025-07-01 16:34:46 +02:00
committed by GitHub
parent 483be9aac2
commit d4ee470b00
268 changed files with 862 additions and 890 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.github/PULL_REQUEST_TEMPLATE.md vendored
View File

@@ -24,7 +24,7 @@ Examples:
pytest -sx tests/test_stuff.py::test_something
```
```bash
python lerobot/scripts/train.py --some.option=true
python -m lerobot.scripts.train --some.option=true
```
## SECTION TO REMOVE BEFORE SUBMITTING YOUR PR

4
.github/workflows/nightly-tests.yml vendored
View File

@@ -44,7 +44,7 @@ jobs:
working-directory: /lerobot
steps:
- name: Tests
run: pytest -v --cov=./lerobot --disable-warnings tests
run: pytest -v --cov=./src/lerobot --disable-warnings tests
- name: Tests end-to-end
run: make test-end-to-end
@@ -74,7 +74,7 @@ jobs:
run: nvidia-smi
- name: Test
run: pytest -v --cov=./lerobot --cov-report=xml --disable-warnings tests
run: pytest -v --cov=./src/lerobot --cov-report=xml --disable-warnings tests
# TODO(aliberts): Link with HF Codecov account
# - name: Upload coverage reports to Codecov with GitHub Action
# uses: codecov/codecov-action@v4

8
.github/workflows/test.yml vendored
View File

@@ -17,7 +17,7 @@ name: Tests
on:
pull_request:
paths:
- "lerobot/**"
- "src/**"
- "tests/**"
- "examples/**"
- ".github/**"
@@ -29,7 +29,7 @@ on:
branches:
- main
paths:
- "lerobot/**"
- "src/**"
- "tests/**"
- "examples/**"
- ".github/**"
@@ -73,7 +73,7 @@ jobs:
- name: Test with pytest
run: |
uv run pytest tests -v --cov=./lerobot --durations=0 \
uv run pytest tests -v --cov=./src/lerobot --durations=0 \
-W ignore::DeprecationWarning:imageio_ffmpeg._utils:7 \
-W ignore::UserWarning:torch.utils.data.dataloader:558 \
-W ignore::UserWarning:gymnasium.utils.env_checker:247 \
@@ -105,7 +105,7 @@ jobs:
- name: Test with pytest
run: |
uv run pytest tests -v --cov=./lerobot --durations=0 \
uv run pytest tests -v --cov=./src/lerobot --durations=0 \
-W ignore::DeprecationWarning:imageio_ffmpeg._utils:7 \
-W ignore::UserWarning:torch.utils.data.dataloader:558 \
-W ignore::UserWarning:gymnasium.utils.env_checker:247 \

2
CONTRIBUTING.md
View File

@@ -67,7 +67,7 @@ post it.
## Adding new policies, datasets or environments
Look at our implementations for [datasets](./lerobot/common/datasets/), [policies](./lerobot/common/policies/),
Look at our implementations for [datasets](./src/lerobot/datasets/), [policies](./src/lerobot/policies/),
environments ([aloha](https://github.com/huggingface/gym-aloha),
[xarm](https://github.com/huggingface/gym-xarm),
[pusht](https://github.com/huggingface/gym-pusht))

4
MANIFEST.in
View File

@@ -1,2 +1,2 @@
include lerobot/templates/lerobot_modelcard_template.md
include lerobot/common/datasets/card_template.md
include src/lerobot/templates/lerobot_modelcard_template.md
include src/lerobot/datasets/card_template.md

18
Makefile
View File

@@ -44,7 +44,7 @@ test-end-to-end:
${MAKE} DEVICE=$(DEVICE) test-smolvla-ete-eval
test-act-ete-train:
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=act \
--policy.dim_model=64 \
--policy.n_action_steps=20 \
@@ -68,12 +68,12 @@ test-act-ete-train:
--output_dir=tests/outputs/act/
test-act-ete-train-resume:
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=tests/outputs/act/checkpoints/000002/pretrained_model/train_config.json \
--resume=true
test-act-ete-eval:
python lerobot/scripts/eval.py \
python -m lerobot.scripts.eval \
--policy.path=tests/outputs/act/checkpoints/000004/pretrained_model \
--policy.device=$(DEVICE) \
--env.type=aloha \
@@ -82,7 +82,7 @@ test-act-ete-eval:
--eval.batch_size=1
test-diffusion-ete-train:
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=diffusion \
--policy.down_dims='[64,128,256]' \
--policy.diffusion_step_embed_dim=32 \
@@ -106,7 +106,7 @@ test-diffusion-ete-train:
--output_dir=tests/outputs/diffusion/
test-diffusion-ete-eval:
python lerobot/scripts/eval.py \
python -m lerobot.scripts.eval \
--policy.path=tests/outputs/diffusion/checkpoints/000002/pretrained_model \
--policy.device=$(DEVICE) \
--env.type=pusht \
@@ -115,7 +115,7 @@ test-diffusion-ete-eval:
--eval.batch_size=1
test-tdmpc-ete-train:
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=tdmpc \
--policy.device=$(DEVICE) \
--policy.push_to_hub=false \
@@ -137,7 +137,7 @@ test-tdmpc-ete-train:
--output_dir=tests/outputs/tdmpc/
test-tdmpc-ete-eval:
python lerobot/scripts/eval.py \
python -m lerobot.scripts.eval \
--policy.path=tests/outputs/tdmpc/checkpoints/000002/pretrained_model \
--policy.device=$(DEVICE) \
--env.type=xarm \
@@ -148,7 +148,7 @@ test-tdmpc-ete-eval:
test-smolvla-ete-train:
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=smolvla \
--policy.n_action_steps=20 \
--policy.chunk_size=20 \
@@ -171,7 +171,7 @@ test-smolvla-ete-train:
--output_dir=tests/outputs/smolvla/
test-smolvla-ete-eval:
python lerobot/scripts/eval.py \
python -m lerobot.scripts.eval \
--policy.path=tests/outputs/smolvla/checkpoints/000004/pretrained_model \
--policy.device=$(DEVICE) \
--env.type=aloha \

42
README.md
View File

@@ -149,44 +149,20 @@ wandb login
(note: you will also need to enable WandB in the configuration. See below.)
## Walkthrough
```
.
├── examples # contains demonstration examples, start here to learn about LeRobot
| └── advanced # contains even more examples for those who have mastered the basics
├── lerobot
| ├── configs # contains config classes with all options that you can override in the command line
| ├── common # contains classes and utilities
| | ├── datasets # various datasets of human demonstrations: aloha, pusht, xarm
| | ├── envs # various sim environments: aloha, pusht, xarm
| | ├── policies # various policies: act, diffusion, tdmpc
| | ├── robot_devices # various real devices: dynamixel motors, opencv cameras, koch robots
| | └── utils # various utilities
| └── scripts # contains functions to execute via command line
| ├── eval.py # load policy and evaluate it on an environment
| ├── train.py # train a policy via imitation learning and/or reinforcement learning
| ├── control_robot.py # teleoperate a real robot, record data, run a policy
| ├── push_dataset_to_hub.py # convert your dataset into LeRobot dataset format and upload it to the Hugging Face hub
| └── visualize_dataset.py # load a dataset and render its demonstrations
├── outputs # contains results of scripts execution: logs, videos, model checkpoints
└── tests # contains pytest utilities for continuous integration
```
### Visualize datasets
Check out [example 1](./examples/1_load_lerobot_dataset.py) that illustrates how to use our dataset class which automatically downloads data from the Hugging Face hub.
You can also locally visualize episodes from a dataset on the hub by executing our script from the command line:
```bash
python lerobot/scripts/visualize_dataset.py \
python -m lerobot.scripts.visualize_dataset \
--repo-id lerobot/pusht \
--episode-index 0
```
or from a dataset in a local folder with the `root` option and the `--local-files-only` (in the following case the dataset will be searched for in `./my_local_data_dir/lerobot/pusht`)
```bash
python lerobot/scripts/visualize_dataset.py \
python -m lerobot.scripts.visualize_dataset \
--repo-id lerobot/pusht \
--root ./my_local_data_dir \
--local-files-only 1 \
@@ -199,7 +175,7 @@ It will open `rerun.io` and display the camera streams, robot states and actions
https://github-production-user-asset-6210df.s3.amazonaws.com/4681518/328035972-fd46b787-b532-47e2-bb6f-fd536a55a7ed.mov?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAVCODYLSA53PQK4ZA%2F20240505%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20240505T172924Z&X-Amz-Expires=300&X-Amz-Signature=d680b26c532eeaf80740f08af3320d22ad0b8a4e4da1bcc4f33142c15b509eda&X-Amz-SignedHeaders=host&actor_id=24889239&key_id=0&repo_id=748713144
Our script can also visualize datasets stored on a distant server. See `python lerobot/scripts/visualize_dataset.py --help` for more instructions.
Our script can also visualize datasets stored on a distant server. See `python -m lerobot.scripts.visualize_dataset --help` for more instructions.
### The `LeRobotDataset` format
@@ -252,7 +228,7 @@ Check out [example 2](./examples/2_evaluate_pretrained_policy.py) that illustrat
We also provide a more capable script to parallelize the evaluation over multiple environments during the same rollout. Here is an example with a pretrained model hosted on [lerobot/diffusion_pusht](https://huggingface.co/lerobot/diffusion_pusht):
```bash
python lerobot/scripts/eval.py \
python -m lerobot.scripts.eval \
--policy.path=lerobot/diffusion_pusht \
--env.type=pusht \
--eval.batch_size=10 \
@@ -264,10 +240,10 @@ python lerobot/scripts/eval.py \
Note: After training your own policy, you can re-evaluate the checkpoints with:
```bash
python lerobot/scripts/eval.py --policy.path={OUTPUT_DIR}/checkpoints/last/pretrained_model
python -m lerobot.scripts.eval --policy.path={OUTPUT_DIR}/checkpoints/last/pretrained_model
```
See `python lerobot/scripts/eval.py --help` for more instructions.
See `python -m lerobot.scripts.eval --help` for more instructions.
### Train your own policy
@@ -279,14 +255,14 @@ A link to the wandb logs for the run will also show up in yellow in your termina
![](media/wandb.png)
Note: For efficiency, during training every checkpoint is evaluated on a low number of episodes. You may use `--eval.n_episodes=500` to evaluate on more episodes than the default. Or, after training, you may want to re-evaluate your best checkpoints on more episodes or change the evaluation settings. See `python lerobot/scripts/eval.py --help` for more instructions.
Note: For efficiency, during training every checkpoint is evaluated on a low number of episodes. You may use `--eval.n_episodes=500` to evaluate on more episodes than the default. Or, after training, you may want to re-evaluate your best checkpoints on more episodes or change the evaluation settings. See `python -m lerobot.scripts.eval --help` for more instructions.
#### Reproduce state-of-the-art (SOTA)
We provide some pretrained policies on our [hub page](https://huggingface.co/lerobot) that can achieve state-of-the-art performances.
You can reproduce their training by loading the config from their run. Simply running:
```bash
python lerobot/scripts/train.py --config_path=lerobot/diffusion_pusht
python -m lerobot.scripts.train --config_path=lerobot/diffusion_pusht
```
reproduces SOTA results for Diffusion Policy on the PushT task.
@@ -312,7 +288,7 @@ python lerobot/scripts/push_dataset_to_hub.py \
See `python lerobot/scripts/push_dataset_to_hub.py --help` for more instructions.
If your dataset format is not supported, implement your own in `lerobot/common/datasets/push_dataset_to_hub/${raw_format}_format.py` by copying examples like [pusht_zarr](https://github.com/huggingface/lerobot/blob/main/lerobot/common/datasets/push_dataset_to_hub/pusht_zarr_format.py), [umi_zarr](https://github.com/huggingface/lerobot/blob/main/lerobot/common/datasets/push_dataset_to_hub/umi_zarr_format.py), [aloha_hdf5](https://github.com/huggingface/lerobot/blob/main/lerobot/common/datasets/push_dataset_to_hub/aloha_hdf5_format.py), or [xarm_pkl](https://github.com/huggingface/lerobot/blob/main/lerobot/common/datasets/push_dataset_to_hub/xarm_pkl_format.py). -->
If your dataset format is not supported, implement your own in `lerobot/datasets/push_dataset_to_hub/${raw_format}_format.py` by copying examples like [pusht_zarr](https://github.com/huggingface/lerobot/blob/main/lerobot/datasets/push_dataset_to_hub/pusht_zarr_format.py), [umi_zarr](https://github.com/huggingface/lerobot/blob/main/lerobot/datasets/push_dataset_to_hub/umi_zarr_format.py), [aloha_hdf5](https://github.com/huggingface/lerobot/blob/main/lerobot/datasets/push_dataset_to_hub/aloha_hdf5_format.py), or [xarm_pkl](https://github.com/huggingface/lerobot/blob/main/lerobot/datasets/push_dataset_to_hub/xarm_pkl_format.py). -->
### Add a pretrained policy

6
benchmarks/video/run_video_benchmark.py
View File

@@ -35,12 +35,12 @@ import torch
from skimage.metrics import mean_squared_error, peak_signal_noise_ratio, structural_similarity
from tqdm import tqdm
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.datasets.video_utils import (
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.video_utils import (
decode_video_frames_torchvision,
encode_video_frames,
)
from lerobot.common.utils.benchmark import TimeBenchmark
from lerobot.utils.benchmark import TimeBenchmark
BASE_ENCODING = OrderedDict(
[

14
docs/source/cameras.mdx
View File

@@ -8,7 +8,7 @@ To instantiate a camera, you need a camera identifier. This identifier might cha
To find the camera indices of the cameras plugged into your system, run the following script:
```bash
python lerobot/find_cameras.py opencv # or realsense for Intel Realsense cameras
python -m lerobot.find_cameras opencv # or realsense for Intel Realsense cameras
```
The output will look something like this if you have two cameras connected:
@@ -44,9 +44,9 @@ Below are two examples, demonstrating how to work with the API.
<hfoption id="Open CV Camera">
```python
from lerobot.common.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.common.cameras.opencv.camera_opencv import OpenCVCamera
from lerobot.common.cameras.configs import ColorMode, Cv2Rotation
from lerobot.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.cameras.opencv.camera_opencv import OpenCVCamera
from lerobot.cameras.configs import ColorMode, Cv2Rotation
# Construct an `OpenCVCameraConfig` with your desired FPS, resolution, color mode, and rotation.
config = OpenCVCameraConfig(
@@ -75,9 +75,9 @@ finally:
<hfoption id="Intel Realsense Camera">
```python
from lerobot.common.cameras.realsense.configuration_realsense import RealSenseCameraConfig
from lerobot.common.cameras.realsense.camera_realsense import RealSenseCamera
from lerobot.common.cameras.configs import ColorMode, Cv2Rotation
from lerobot.cameras.realsense.configuration_realsense import RealSenseCameraConfig
from lerobot.cameras.realsense.camera_realsense import RealSenseCamera
from lerobot.cameras.configs import ColorMode, Cv2Rotation
# Create a `RealSenseCameraConfig` specifying your camera’s serial number and enabling depth.
config = RealSenseCameraConfig(

30
docs/source/hilserl.mdx
View File

@@ -50,12 +50,12 @@ pip install -e ".[hilserl]"
### Understanding Configuration
The training process begins with proper configuration for the HILSerl environment. The configuration class of interest is `HILSerlRobotEnvConfig` in `lerobot/common/envs/configs.py`. Which is defined as:
The training process begins with proper configuration for the HILSerl environment. The configuration class of interest is `HILSerlRobotEnvConfig` in `lerobot/envs/configs.py`. Which is defined as:
```python
class HILSerlRobotEnvConfig(EnvConfig):
robot: RobotConfig | None = None # Main robot agent (defined in `lerobot/common/robots`)
teleop: TeleoperatorConfig | None = None # Teleoperator agent, e.g., gamepad or leader arm, (defined in `lerobot/common/teleoperators`)
robot: RobotConfig | None = None # Main robot agent (defined in `lerobot/robots`)
teleop: TeleoperatorConfig | None = None # Teleoperator agent, e.g., gamepad or leader arm, (defined in `lerobot/teleoperators`)
wrapper: EnvTransformConfig | None = None # Environment wrapper settings; check `lerobot/scripts/server/gym_manipulator.py`
fps: int = 10 # Control frequency
name: str = "real_robot" # Environment name
@@ -172,7 +172,7 @@ class SO100FollowerEndEffectorConfig(SO100FollowerConfig):
)
```
The `Teleoperator` defines the teleoperation device. You can check the list of available teleoperators in `lerobot/common/teleoperators`.
The `Teleoperator` defines the teleoperation device. You can check the list of available teleoperators in `lerobot/teleoperators`.
**Setting up the Gamepad**
@@ -226,7 +226,7 @@ During the online training, press `space` to take over the policy and `space` ag
Start the recording process, an example of the config file can be found [here](https://huggingface.co/datasets/aractingi/lerobot-example-config-files/blob/main/env_config_so100.json):
```bash
python lerobot/scripts/rl/gym_manipulator.py --config_path lerobot/configs/env_config_so100.json
python -m lerobot.scripts.rl.gym_manipulator --config_path src/lerobot/configs/env_config_so100.json
```
During recording:
@@ -256,7 +256,7 @@ Note: If you already know the crop parameters, you can skip this step and just s
Use the `crop_dataset_roi.py` script to interactively select regions of interest in your camera images:
```bash
python lerobot/scripts/rl/crop_dataset_roi.py --repo-id username/pick_lift_cube
python -m lerobot.scripts.rl.crop_dataset_roi --repo-id username/pick_lift_cube
```
1. For each camera view, the script will display the first frame
@@ -313,7 +313,7 @@ Before training, you need to collect a dataset with labeled examples. The `recor
To collect a dataset, you need to modify some parameters in the environment configuration based on HILSerlRobotEnvConfig.
```bash
python lerobot/scripts/rl/gym_manipulator.py --config_path lerobot/configs/reward_classifier_train_config.json
python -m lerobot.scripts.rl.gym_manipulator --config_path src/lerobot/configs/reward_classifier_train_config.json
```
**Key Parameters for Data Collection**
@@ -387,7 +387,7 @@ Example configuration for training the [reward classifier](https://huggingface.c
To train the classifier, use the `train.py` script with your configuration:
```bash
python lerobot/scripts/train.py --config_path path/to/reward_classifier_train_config.json
python -m lerobot.scripts.train --config_path path/to/reward_classifier_train_config.json
```
**Deploying and Testing the Model**
@@ -410,7 +410,7 @@ or set the argument in the json config file.
Run `gym_manipulator.py` to test the model.
```bash
python lerobot/scripts/rl/gym_manipulator.py --config_path path/to/env_config.json
python -m lerobot.scripts.rl.gym_manipulator --config_path path/to/env_config.json
```
The reward classifier will automatically provide rewards based on the visual input from the robot's cameras.
@@ -422,17 +422,17 @@ The reward classifier will automatically provide rewards based on the visual inp
2. **Collect a dataset**:
```bash
python lerobot/scripts/rl/gym_manipulator.py --config_path lerobot/configs/env_config.json
python -m lerobot.scripts.rl.gym_manipulator --config_path src/lerobot/configs/env_config.json
```
3. **Train the classifier**:
```bash
python lerobot/scripts/train.py --config_path lerobot/configs/reward_classifier_train_config.json
python -m lerobot.scripts.train --config_path src/lerobot/configs/reward_classifier_train_config.json
```
4. **Test the classifier**:
```bash
python lerobot/scripts/rl/gym_manipulator.py --config_path lerobot/configs/env_config.json
python -m lerobot.scripts.rl.gym_manipulator --config_path src/lerobot/configs/env_config.json
```
### Training with Actor-Learner
@@ -446,7 +446,7 @@ Create a training configuration file (example available [here](https://huggingfa
1. Configure the policy settings (`type="sac"`, `device`, etc.)
2. Set `dataset` to your cropped dataset
3. Configure environment settings with crop parameters
4. Check the other parameters related to SAC in [configuration_sac.py](https://github.com/huggingface/lerobot/blob/19bb621a7d0a31c20cd3cc08b1dbab68d3031454/lerobot/common/policies/sac/configuration_sac.py#L79).
4. Check the other parameters related to SAC in [configuration_sac.py](https://github.com/huggingface/lerobot/blob/19bb621a7d0a31c20cd3cc08b1dbab68d3031454/lerobot/policies/sac/configuration_sac.py#L79).
5. Verify that the `policy` config is correct with the right `input_features` and `output_features` for your task.
**Starting the Learner**
@@ -454,7 +454,7 @@ Create a training configuration file (example available [here](https://huggingfa
First, start the learner server process:
```bash
python lerobot/scripts/rl/learner.py --config_path lerobot/configs/train_config_hilserl_so100.json
python -m lerobot.scripts.rl.learner --config_path src/lerobot/configs/train_config_hilserl_so100.json
```
The learner:
@@ -468,7 +468,7 @@ The learner:
In a separate terminal, start the actor process with the same configuration:
```bash
python lerobot/scripts/rl/actor.py --config_path lerobot/configs/train_config_hilserl_so100.json
python -m lerobot.scripts.rl.actor --config_path src/lerobot/configs/train_config_hilserl_so100.json
```
The actor:

8
docs/source/hilserl_sim.mdx
View File

@@ -77,7 +77,7 @@ Important parameters:
To run the environment, set mode to null:
```python
python lerobot/scripts/rl/gym_manipulator.py --config_path path/to/gym_hil_env.json
python -m lerobot.scripts.rl.gym_manipulator --config_path path/to/gym_hil_env.json
```
### Recording a Dataset
@@ -85,7 +85,7 @@ python lerobot/scripts/rl/gym_manipulator.py --config_path path/to/gym_hil_env.j
To collect a dataset, set the mode to `record` whilst defining the repo_id and number of episodes to record:
```python
python lerobot/scripts/rl/gym_manipulator.py --config_path path/to/gym_hil_env.json
python -m lerobot.scripts.rl.gym_manipulator --config_path path/to/gym_hil_env.json
```
### Training a Policy
@@ -93,13 +93,13 @@ python lerobot/scripts/rl/gym_manipulator.py --config_path path/to/gym_hil_env.j
To train a policy, checkout the configuration example available [here](https://huggingface.co/datasets/aractingi/lerobot-example-config-files/blob/main/train_gym_hil_env.json) and run the actor and learner servers:
```python
python lerobot/scripts/rl/actor.py --config_path path/to/train_gym_hil_env.json
python -m lerobot.scripts.rl.actor --config_path path/to/train_gym_hil_env.json
```
In a different terminal, run the learner server:
```python
python lerobot/scripts/rl/learner.py --config_path path/to/train_gym_hil_env.json
python -m lerobot.scripts.rl.learner --config_path path/to/train_gym_hil_env.json
```
The simulation environment provides a safe and repeatable way to develop and test your Human-In-the-Loop reinforcement learning components before deploying to real robots.

64
docs/source/il_robots.mdx
View File

@@ -52,8 +52,8 @@ python -m lerobot.teleoperate \
</hfoption>
<hfoption id="API example">
```python
from lerobot.common.teleoperators.so101_leader import SO101LeaderConfig, SO101Leader
from lerobot.common.robots.so101_follower import SO101FollowerConfig, SO101Follower
from lerobot.teleoperators.so101_leader import SO101LeaderConfig, SO101Leader
from lerobot.robots.so101_follower import SO101FollowerConfig, SO101Follower
robot_config = SO101FollowerConfig(
port="/dev/tty.usbmodem58760431541",
@@ -105,9 +105,9 @@ python -m lerobot.teleoperate \
</hfoption>
<hfoption id="API example">
```python
from lerobot.common.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.common.teleoperators.koch_leader import KochLeaderConfig, KochLeader
from lerobot.common.robots.koch_follower import KochFollowerConfig, KochFollower
from lerobot.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.teleoperators.koch_leader import KochLeaderConfig, KochLeader
from lerobot.robots.koch_follower import KochFollowerConfig, KochFollower
camera_config = {
"front": OpenCVCameraConfig(index_or_path=0, width=1920, height=1080, fps=30)
@@ -175,15 +175,15 @@ python -m lerobot.record \
</hfoption>
<hfoption id="API example">
```python
from lerobot.common.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.datasets.utils import hw_to_dataset_features
from lerobot.common.robots.so100_follower import SO100Follower, SO100FollowerConfig
from lerobot.common.teleoperators.so100_leader.config_so100_leader import SO100LeaderConfig
from lerobot.common.teleoperators.so100_leader.so100_leader import SO100Leader
from lerobot.common.utils.control_utils import init_keyboard_listener
from lerobot.common.utils.utils import log_say
from lerobot.common.utils.visualization_utils import _init_rerun
from lerobot.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.utils import hw_to_dataset_features
from lerobot.robots.so100_follower import SO100Follower, SO100FollowerConfig
from lerobot.teleoperators.so100_leader.config_so100_leader import SO100LeaderConfig
from lerobot.teleoperators.so100_leader.so100_leader import SO100Leader
from lerobot.utils.control_utils import init_keyboard_listener
from lerobot.utils.utils import log_say
from lerobot.utils.visualization_utils import _init_rerun
from lerobot.record import record_loop
NUM_EPISODES = 5
@@ -353,11 +353,11 @@ python -m lerobot.replay \
```python
import time
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.robots.so100_follower.config_so100_follower import SO100FollowerConfig
from lerobot.common.robots.so100_follower.so100_follower import SO100Follower
from lerobot.common.utils.robot_utils import busy_wait
from lerobot.common.utils.utils import log_say
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.robots.so100_follower.config_so100_follower import SO100FollowerConfig
from lerobot.robots.so100_follower.so100_follower import SO100Follower
from lerobot.utils.robot_utils import busy_wait
from lerobot.utils.utils import log_say
episode_idx = 0
@@ -389,9 +389,9 @@ Your robot should replicate movements similar to those you recorded. For example
## Train a policy
To train a policy to control your robot, use the [`python lerobot/scripts/train.py`](../lerobot/scripts/train.py) script. A few arguments are required. Here is an example command:
To train a policy to control your robot, use the [`python -m lerobot.scripts.train`](../src/lerobot/scripts/train.py) script. A few arguments are required. Here is an example command:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--dataset.repo_id=${HF_USER}/so101_test \
--policy.type=act \
--output_dir=outputs/train/act_so101_test \
@@ -403,7 +403,7 @@ python lerobot/scripts/train.py \
Let's explain the command:
1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/so101_test`.
2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor states, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../src/lerobot/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor states, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
@@ -411,7 +411,7 @@ Training should take several hours. You will find checkpoints in `outputs/train/
To resume training from a checkpoint, below is an example command to resume from `last` checkpoint of the `act_so101_test` policy:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=outputs/train/act_so101_test/checkpoints/last/pretrained_model/train_config.json \
--resume=true
```
@@ -462,15 +462,15 @@ python -m lerobot.record \
</hfoption>
<hfoption id="API example">
```python
from lerobot.common.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.datasets.utils import hw_to_dataset_features
from lerobot.common.policies.act.modeling_act import ACTPolicy
from lerobot.common.robots.so100_follower.config_so100_follower import SO100FollowerConfig
from lerobot.common.robots.so100_follower.so100_follower import SO100Follower
from lerobot.common.utils.control_utils import init_keyboard_listener
from lerobot.common.utils.utils import log_say
from lerobot.common.utils.visualization_utils import _init_rerun
from lerobot.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.utils import hw_to_dataset_features
from lerobot.policies.act.modeling_act import ACTPolicy
from lerobot.robots.so100_follower.config_so100_follower import SO100FollowerConfig
from lerobot.robots.so100_follower.so100_follower import SO100Follower
from lerobot.utils.control_utils import init_keyboard_listener
from lerobot.utils.utils import log_say
from lerobot.utils.visualization_utils import _init_rerun
from lerobot.record import record_loop
NUM_EPISODES = 5

14
docs/source/il_sim.mdx
View File

@@ -35,14 +35,14 @@ Then we can run this command to start:
<hfoption id="Linux">
```bash
python lerobot/scripts/rl/gym_manipulator.py --config_path path/to/env_config_gym_hil_il.json
python -m lerobot.scripts.rl.gym_manipulator --config_path path/to/env_config_gym_hil_il.json
```
</hfoption>
<hfoption id="MacOS">
```bash
mjpython lerobot/scripts/rl/gym_manipulator.py --config_path path/to/env_config_gym_hil_il.json
mjpython -m lerobot.scripts.rl.gym_manipulator --config_path path/to/env_config_gym_hil_il.json
```
</hfoption>
@@ -81,9 +81,9 @@ If you uploaded your dataset to the hub you can [visualize your dataset online](
## Train a policy
To train a policy to control your robot, use the [`python lerobot/scripts/train.py`](../lerobot/scripts/train.py) script. A few arguments are required. Here is an example command:
To train a policy to control your robot, use the [`python -m lerobot.scripts.train`](../src/lerobot/scripts/train.py) script. A few arguments are required. Here is an example command:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--dataset.repo_id=${HF_USER}/il_gym \
--policy.type=act \
--output_dir=outputs/train/il_sim_test \
@@ -94,7 +94,7 @@ python lerobot/scripts/train.py \
Let's explain the command:
1. We provided the dataset as argument with `--dataset.repo_id=${HF_USER}/il_gym`.
2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../lerobot/common/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor states, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
2. We provided the policy with `policy.type=act`. This loads configurations from [`configuration_act.py`](../src/lerobot/policies/act/configuration_act.py). Importantly, this policy will automatically adapt to the number of motor states, motor actions and cameras of your robot (e.g. `laptop` and `phone`) which have been saved in your dataset.
4. We provided `policy.device=cuda` since we are training on a Nvidia GPU, but you could use `policy.device=mps` to train on Apple silicon.
5. We provided `wandb.enable=true` to use [Weights and Biases](https://docs.wandb.ai/quickstart) for visualizing training plots. This is optional but if you use it, make sure you are logged in by running `wandb login`.
@@ -130,14 +130,14 @@ Then you can run this command to visualize your trained policy
<hfoption id="Linux">
```bash
python lerobot/scripts/rl/eval_policy.py --config_path=path/to/eval_config_gym_hil.json
python -m lerobot.scripts.rl.eval_policy --config_path=path/to/eval_config_gym_hil.json
```
</hfoption>
<hfoption id="MacOS">
```bash
mjpython lerobot/scripts/rl/eval_policy.py --config_path=path/to/eval_config_gym_hil.json
mjpython -m lerobot.scripts.rl.eval_policy --config_path=path/to/eval_config_gym_hil.json
```
</hfoption>

28
docs/source/integrate_hardware.mdx
View File

@@ -2,7 +2,7 @@
This tutorial will explain how to integrate your own robot design into the LeRobot ecosystem and have it access all of our tools (data collection, control pipelines, policy training and inference).
To that end, we provide the [`Robot`](https://github.com/huggingface/lerobot/blob/main/lerobot/common/robots/robot.py) base class in the LeRobot which specifies a standard interface for physical robot integration. Let's see how to implement it.
To that end, we provide the [`Robot`](https://github.com/huggingface/lerobot/blob/main/lerobot/robots/robot.py) base class in the LeRobot which specifies a standard interface for physical robot integration. Let's see how to implement it.
## Prerequisites
@@ -14,11 +14,11 @@ To that end, we provide the [`Robot`](https://github.com/huggingface/lerobot/blo
If you're using Feetech or Dynamixel motors, LeRobot provides built-in bus interfaces:
- [`FeetechMotorsBus`](https://github.com/huggingface/lerobot/blob/main/lerobot/common/motors/feetech/feetech.py) – for controlling Feetech servos
- [`DynamixelMotorsBus`](https://github.com/huggingface/lerobot/blob/main/lerobot/common/motors/dynamixel/dynamixel.py) – for controlling Dynamixel servos
- [`FeetechMotorsBus`](https://github.com/huggingface/lerobot/blob/main/lerobot/motors/feetech/feetech.py) – for controlling Feetech servos
- [`DynamixelMotorsBus`](https://github.com/huggingface/lerobot/blob/main/lerobot/motors/dynamixel/dynamixel.py) – for controlling Dynamixel servos
Please refer to the [`MotorsBus`](https://github.com/huggingface/lerobot/blob/main/lerobot/common/motors/motors_bus.py) abstract class to learn about its API.
For a good example of how it can be used, you can have a look at our own [SO101 follower implementation](https://github.com/huggingface/lerobot/blob/main/lerobot/common/robots/so101_follower/so101_follower.py)
Please refer to the [`MotorsBus`](https://github.com/huggingface/lerobot/blob/main/lerobot/motors/motors_bus.py) abstract class to learn about its API.
For a good example of how it can be used, you can have a look at our own [SO101 follower implementation](https://github.com/huggingface/lerobot/blob/main/lerobot/robots/so101_follower/so101_follower.py)
Use these if compatible. Otherwise, you'll need to find or write a Python interface (not covered in this tutorial):
- Find an existing SDK in Python (or use bindings to C/C++)
@@ -32,7 +32,7 @@ For Feetech and Dynamixel, we currently support these servos:
- SCS series (protocol 1): `scs0009`
- Dynamixel (protocol 2.0 only): `xl330-m077`, `xl330-m288`, `xl430-w250`, `xm430-w350`, `xm540-w270`, `xc430-w150`
If you are using Feetech or Dynamixel servos that are not in this list, you can add those in the [Feetech table](https://github.com/huggingface/lerobot/blob/main/lerobot/common/motors/feetech/tables.py) or [Dynamixel table](https://github.com/huggingface/lerobot/blob/main/lerobot/common/motors/dynamixel/tables.py). Depending on the model, this will require you to add model-specific information. In most cases though, there shouldn't be a lot of additions to do.
If you are using Feetech or Dynamixel servos that are not in this list, you can add those in the [Feetech table](https://github.com/huggingface/lerobot/blob/main/lerobot/motors/feetech/tables.py) or [Dynamixel table](https://github.com/huggingface/lerobot/blob/main/lerobot/motors/dynamixel/tables.py). Depending on the model, this will require you to add model-specific information. In most cases though, there shouldn't be a lot of additions to do.
In the next sections, we'll use a `FeetechMotorsBus` as the motors interface for the examples. Replace it and adapt to your motors if necessary.
@@ -44,9 +44,9 @@ Here, we'll add the port name and one camera by default for our robot:
```python
from dataclasses import dataclass, field
from lerobot.common.cameras import CameraConfig
from lerobot.common.cameras.opencv import OpenCVCameraConfig
from lerobot.common.robots import RobotConfig
from lerobot.cameras import CameraConfig
from lerobot.cameras.opencv import OpenCVCameraConfig
from lerobot.robots import RobotConfig
@RobotConfig.register_subclass("my_cool_robot")
@@ -72,10 +72,10 @@ Next, we'll create our actual robot class which inherits from `Robot`. This abst
Here we'll create a simple 5-DoF robot with one camera. It could be a simple arm but notice that the `Robot` abstract class does not assume anything on your robot's form factor. You can let you imagination run wild when designing new robots!
```python
from lerobot.common.cameras import make_cameras_from_configs
from lerobot.common.motors import Motor, MotorNormMode
from lerobot.common.motors.feetech import FeetechMotorsBus
from lerobot.common.robots import Robot
from lerobot.cameras import make_cameras_from_configs
from lerobot.motors import Motor, MotorNormMode
from lerobot.motors.feetech import FeetechMotorsBus
from lerobot.robots import Robot
class MyCoolRobot(Robot):
config_class = MyCoolRobotConfig
@@ -303,7 +303,7 @@ def send_action(self, action: dict[str, Any]) -> dict[str, Any]:
## Adding a Teleoperator
For implementing teleoperation devices, we also provide a [`Teleoperator`](https://github.com/huggingface/lerobot/blob/main/lerobot/common/teleoperators/teleoperator.py) base class. This class is very similar to the `Robot` base class and also doesn't assume anything on form factor.
For implementing teleoperation devices, we also provide a [`Teleoperator`](https://github.com/huggingface/lerobot/blob/main/lerobot/teleoperators/teleoperator.py) base class. This class is very similar to the `Robot` base class and also doesn't assume anything on form factor.
The main differences are in the I/O functions: a teleoperator allows you to produce action via `get_action` and can receive feedback actions via `send_feedback`. Feedback could be anything controllable on the teleoperation device that could help the person controlling it understand the consequences of the actions sent. Think motion/force feedback on a leader arm, vibrations on a gamepad controller for example. To implement a teleoperator, you can follow this same tutorial and adapt it for these two methods.

2
docs/source/koch.mdx
View File

@@ -1 +1 @@
../../lerobot/common/robots/koch_follower/koch.mdx
../../src/lerobot/robots/koch_follower/koch.mdx

2
docs/source/lekiwi.mdx
View File

@@ -1 +1 @@
../../lerobot/common/robots/lekiwi/lekiwi.mdx
../../src/lerobot/robots/lekiwi/lekiwi.mdx

4
docs/source/smolvla.mdx
View File

@@ -44,7 +44,7 @@ If you don't have a gpu device, you can train using our notebook on [![Google Co
Pass your dataset to the training script using `--dataset.repo_id`. If you want to test your installation, run the following command where we use one of the datasets we collected for the [SmolVLA Paper](https://huggingface.co/papers/2506.01844).
```bash
cd lerobot && python lerobot/scripts/train.py \
cd lerobot && python -m lerobot.scripts.train \
--policy.path=lerobot/smolvla_base \
--dataset.repo_id=${HF_USER}/mydataset \
--batch_size=64 \
@@ -62,7 +62,7 @@ You can start with a small batch size and increase it incrementally, if the GPU
Fine-tuning is an art. For a complete overview of the options for finetuning, run
```bash
python lerobot/scripts/train.py --help
python -m lerobot.scripts.train --help
```
<p align="center">

2
docs/source/so100.mdx
View File

@@ -1 +1 @@
../../lerobot/common/robots/so100_follower/so100.mdx
../../src/lerobot/robots/so100_follower/so100.mdx

2
docs/source/so101.mdx
View File

@@ -1 +1 @@
../../lerobot/common/robots/so101_follower/so101.mdx
../../src/lerobot/robots/so101_follower/so101.mdx

2
examples/1_load_lerobot_dataset.py
View File

@@ -32,7 +32,7 @@ import torch
from huggingface_hub import HfApi
import lerobot
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset, LeRobotDatasetMetadata
from lerobot.datasets.lerobot_dataset import LeRobotDataset, LeRobotDatasetMetadata
# We ported a number of existing datasets ourselves, use this to see the list:
print("List of available datasets:")

2
examples/2_evaluate_pretrained_policy.py
View File

@@ -30,7 +30,7 @@ import imageio
import numpy
import torch
from lerobot.common.policies.diffusion.modeling_diffusion import DiffusionPolicy
from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
# Create a directory to store the video of the evaluation
output_directory = Path("outputs/eval/example_pusht_diffusion")

8
examples/3_train_policy.py
View File

@@ -22,11 +22,11 @@ from pathlib import Path
import torch
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset, LeRobotDatasetMetadata
from lerobot.common.datasets.utils import dataset_to_policy_features
from lerobot.common.policies.diffusion.configuration_diffusion import DiffusionConfig
from lerobot.common.policies.diffusion.modeling_diffusion import DiffusionPolicy
from lerobot.configs.types import FeatureType
from lerobot.datasets.lerobot_dataset import LeRobotDataset, LeRobotDatasetMetadata
from lerobot.datasets.utils import dataset_to_policy_features
from lerobot.policies.diffusion.configuration_diffusion import DiffusionConfig
from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
def main():

42
examples/4_train_policy_with_script.md
View File

@@ -4,7 +4,7 @@ This tutorial will explain the training script, how to use it, and particularly
## The training script
LeRobot offers a training script at [`lerobot/scripts/train.py`](../lerobot/scripts/train.py). At a high level it does the following:
LeRobot offers a training script at [`lerobot/scripts/train.py`](../src/lerobot/scripts/train.py). At a high level it does the following:
- Initialize/load a configuration for the following steps using.
- Instantiates a dataset.
@@ -21,7 +21,7 @@ In the training script, the main function `train` expects a `TrainPipelineConfig
def train(cfg: TrainPipelineConfig):
```
You can inspect the `TrainPipelineConfig` defined in [`lerobot/configs/train.py`](../lerobot/configs/train.py) (which is heavily commented and meant to be a reference to understand any option)
You can inspect the `TrainPipelineConfig` defined in [`lerobot/configs/train.py`](../src/lerobot/configs/train.py) (which is heavily commented and meant to be a reference to understand any option)
When running the script, inputs for the command line are parsed thanks to the `@parser.wrap()` decorator and an instance of this class is automatically generated. Under the hood, this is done with [Draccus](https://github.com/dlwh/draccus) which is a tool dedicated to this purpose. If you're familiar with Hydra, Draccus can similarly load configurations from config files (.json, .yaml) and also override their values through command line inputs. Unlike Hydra, these configurations are pre-defined in the code through dataclasses rather than being defined entirely in config files. This allows for more rigorous serialization/deserialization, typing, and to manipulate configuration as objects directly in the code and not as dictionaries or namespaces (which enables nice features in an IDE such as autocomplete, jump-to-def, etc.)
@@ -50,9 +50,9 @@ By default, every field takes its default value specified in the dataclass. If a
## Specifying values from the CLI
Let's say that we want to train [Diffusion Policy](../lerobot/common/policies/diffusion) on the [pusht](https://huggingface.co/datasets/lerobot/pusht) dataset, using the [gym_pusht](https://github.com/huggingface/gym-pusht) environment for evaluation. The command to do so would look like this:
Let's say that we want to train [Diffusion Policy](../src/lerobot/policies/diffusion) on the [pusht](https://huggingface.co/datasets/lerobot/pusht) dataset, using the [gym_pusht](https://github.com/huggingface/gym-pusht) environment for evaluation. The command to do so would look like this:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--dataset.repo_id=lerobot/pusht \
--policy.type=diffusion \
--env.type=pusht
@@ -60,12 +60,12 @@ python lerobot/scripts/train.py \
Let's break this down:
- To specify the dataset, we just need to specify its `repo_id` on the hub which is the only required argument in the `DatasetConfig`. The rest of the fields have default values and in this case we are fine with those so we can just add the option `--dataset.repo_id=lerobot/pusht`.
- To specify the policy, we can just select diffusion policy using `--policy` appended with `.type`. Here, `.type` is a special argument which allows us to select config classes inheriting from `draccus.ChoiceRegistry` and that have been decorated with the `register_subclass()` method. To have a better explanation of this feature, have a look at this [Draccus demo](https://github.com/dlwh/draccus?tab=readme-ov-file#more-flexible-configuration-with-choice-types). In our code, we use this mechanism mainly to select policies, environments, robots, and some other components like optimizers. The policies available to select are located in [lerobot/common/policies](../lerobot/common/policies)
- Similarly, we select the environment with `--env.type=pusht`. The different environment configs are available in [`lerobot/common/envs/configs.py`](../lerobot/common/envs/configs.py)
- To specify the policy, we can just select diffusion policy using `--policy` appended with `.type`. Here, `.type` is a special argument which allows us to select config classes inheriting from `draccus.ChoiceRegistry` and that have been decorated with the `register_subclass()` method. To have a better explanation of this feature, have a look at this [Draccus demo](https://github.com/dlwh/draccus?tab=readme-ov-file#more-flexible-configuration-with-choice-types). In our code, we use this mechanism mainly to select policies, environments, robots, and some other components like optimizers. The policies available to select are located in [lerobot/policies](../src/lerobot/policies)
- Similarly, we select the environment with `--env.type=pusht`. The different environment configs are available in [`lerobot/envs/configs.py`](../src/lerobot/envs/configs.py)
Let's see another example. Let's say you've been training [ACT](../lerobot/common/policies/act) on [lerobot/aloha_sim_insertion_human](https://huggingface.co/datasets/lerobot/aloha_sim_insertion_human) using the [gym-aloha](https://github.com/huggingface/gym-aloha) environment for evaluation with:
Let's see another example. Let's say you've been training [ACT](../src/lerobot/policies/act) on [lerobot/aloha_sim_insertion_human](https://huggingface.co/datasets/lerobot/aloha_sim_insertion_human) using the [gym-aloha](https://github.com/huggingface/gym-aloha) environment for evaluation with:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=act \
--dataset.repo_id=lerobot/aloha_sim_insertion_human \
--env.type=aloha \
@@ -74,9 +74,9 @@ python lerobot/scripts/train.py \
> Notice we added `--output_dir` to explicitly tell where to write outputs from this run (checkpoints, training state, configs etc.). This is not mandatory and if you don't specify it, a default directory will be created from the current date and time, env.type and policy.type. This will typically look like `outputs/train/2025-01-24/16-10-05_aloha_act`.
We now want to train a different policy for aloha on another task. We'll change the dataset and use [lerobot/aloha_sim_transfer_cube_human](https://huggingface.co/datasets/lerobot/aloha_sim_transfer_cube_human) instead. Of course, we also need to change the task of the environment as well to match this other task.
Looking at the [`AlohaEnv`](../lerobot/common/envs/configs.py) config, the task is `"AlohaInsertion-v0"` by default, which corresponds to the task we trained on in the command above. The [gym-aloha](https://github.com/huggingface/gym-aloha?tab=readme-ov-file#description) environment also has the `AlohaTransferCube-v0` task which corresponds to this other task we want to train on. Putting this together, we can train this new policy on this different task using:
Looking at the [`AlohaEnv`](../src/lerobot/envs/configs.py) config, the task is `"AlohaInsertion-v0"` by default, which corresponds to the task we trained on in the command above. The [gym-aloha](https://github.com/huggingface/gym-aloha?tab=readme-ov-file#description) environment also has the `AlohaTransferCube-v0` task which corresponds to this other task we want to train on. Putting this together, we can train this new policy on this different task using:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=act \
--dataset.repo_id=lerobot/aloha_sim_transfer_cube_human \
--env.type=aloha \
@@ -111,7 +111,7 @@ Now, let's assume that we want to reproduce the run just above. That run has pro
We can then simply load the config values from this file using:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=outputs/train/act_aloha_transfer/checkpoints/last/pretrained_model/ \
--output_dir=outputs/train/act_aloha_transfer_2
```
@@ -119,7 +119,7 @@ python lerobot/scripts/train.py \
Similarly to Hydra, we can still override some parameters in the CLI if we want to, e.g.:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=outputs/train/act_aloha_transfer/checkpoints/last/pretrained_model/ \
--output_dir=outputs/train/act_aloha_transfer_2
--policy.n_action_steps=80
@@ -128,7 +128,7 @@ python lerobot/scripts/train.py \
`--config_path` can also accept the repo_id of a repo on the hub that contains a `train_config.json` file, e.g. running:
```bash
python lerobot/scripts/train.py --config_path=lerobot/diffusion_pusht
python -m lerobot.scripts.train --config_path=lerobot/diffusion_pusht
```
will start a training run with the same configuration used for training [lerobot/diffusion_pusht](https://huggingface.co/lerobot/diffusion_pusht)
@@ -139,7 +139,7 @@ Being able to resume a training run is important in case it crashed or aborted f
Let's reuse the command from the previous run and add a few more options:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=act \
--dataset.repo_id=lerobot/aloha_sim_transfer_cube_human \
--env.type=aloha \
@@ -155,7 +155,7 @@ INFO 2025-01-24 16:10:56 ts/train.py:263 Checkpoint policy after step 100
```
Now let's simulate a crash by killing the process (hit `ctrl`+`c`). We can then simply resume this run from the last checkpoint available with:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=outputs/train/run_resumption/checkpoints/last/pretrained_model/ \
--resume=true
```
@@ -164,7 +164,7 @@ You should see from the logging that your training picks up from where it left o
Another reason for which you might want to resume a run is simply to extend training and add more training steps. The number of training steps is set by the option `--steps`, which is 100 000 by default.
You could double the number of steps of the previous run with:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=outputs/train/run_resumption/checkpoints/last/pretrained_model/ \
--resume=true \
--steps=200000
@@ -195,7 +195,7 @@ In addition to the features currently in Draccus, we've added a special `.path`
For example, we could fine-tune a [policy pre-trained on the aloha transfer task](https://huggingface.co/lerobot/act_aloha_sim_transfer_cube_human) on the aloha insertion task. We can achieve this with:
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.path=lerobot/act_aloha_sim_transfer_cube_human \
--dataset.repo_id=lerobot/aloha_sim_insertion_human \
--env.type=aloha \
@@ -236,7 +236,7 @@ We'll summarize here the main use cases to remember from this tutorial.
#### Train a policy from scratch – CLI
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.type=act \ # <- select 'act' policy
--env.type=pusht \ # <- select 'pusht' environment
--dataset.repo_id=lerobot/pusht # <- train on this dataset
@@ -244,14 +244,14 @@ python lerobot/scripts/train.py \
#### Train a policy from scratch - config file + CLI
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=path/to/pretrained_model \ # <- can also be a repo_id
--policy.n_action_steps=80 # <- you may still override values
```
#### Resume/continue a training run
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--config_path=checkpoint/pretrained_model/ \
--resume=true \
--steps=200000 # <- you can change some training parameters
@@ -259,7 +259,7 @@ python lerobot/scripts/train.py \
#### Fine-tuning
```bash
python lerobot/scripts/train.py \
python -m lerobot.scripts.train \
--policy.path=lerobot/act_aloha_sim_transfer_cube_human \ # <- can also be a local path to a checkpoint
--dataset.repo_id=lerobot/aloha_sim_insertion_human \
--env.type=aloha \

2
examples/advanced/1_add_image_transforms.py
View File

@@ -22,7 +22,7 @@ from pathlib import Path
from torchvision.transforms import ToPILImage, v2
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.lerobot_dataset import LeRobotDataset
dataset_repo_id = "lerobot/aloha_static_screw_driver"

4
examples/advanced/2_calculate_validation_loss.py
View File

@@ -26,8 +26,8 @@ import math
import torch
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset, LeRobotDatasetMetadata
from lerobot.common.policies.diffusion.modeling_diffusion import DiffusionPolicy
from lerobot.datasets.lerobot_dataset import LeRobotDataset, LeRobotDatasetMetadata
from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
def main():

8
examples/backward_compatibility/replay.py
View File

@@ -35,8 +35,8 @@ from pprint import pformat
import draccus
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.robots import ( # noqa: F401
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.robots import ( # noqa: F401
Robot,
RobotConfig,
koch_follower,
@@ -44,8 +44,8 @@ from lerobot.common.robots import ( # noqa: F401
so100_follower,
so101_follower,
)
from lerobot.common.utils.robot_utils import busy_wait
from lerobot.common.utils.utils import (
from lerobot.utils.robot_utils import busy_wait
from lerobot.utils.utils import (
init_logging,
log_say,
)

10
examples/lekiwi/evaluate.py
View File

@@ -1,8 +1,8 @@
from lerobot.common.datasets.utils import build_dataset_frame, hw_to_dataset_features
from lerobot.common.policies.act.modeling_act import ACTPolicy
from lerobot.common.robots.lekiwi import LeKiwiClient, LeKiwiClientConfig
from lerobot.common.utils.control_utils import predict_action
from lerobot.common.utils.utils import get_safe_torch_device
from lerobot.datasets.utils import build_dataset_frame, hw_to_dataset_features
from lerobot.policies.act.modeling_act import ACTPolicy
from lerobot.robots.lekiwi import LeKiwiClient, LeKiwiClientConfig
from lerobot.utils.control_utils import predict_action
from lerobot.utils.utils import get_safe_torch_device
NB_CYCLES_CLIENT_CONNECTION = 1000

12
examples/lekiwi/record.py
View File

@@ -1,11 +1,11 @@
import time
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.datasets.utils import hw_to_dataset_features
from lerobot.common.robots.lekiwi.config_lekiwi import LeKiwiClientConfig
from lerobot.common.robots.lekiwi.lekiwi_client import LeKiwiClient
from lerobot.common.teleoperators.keyboard import KeyboardTeleop, KeyboardTeleopConfig
from lerobot.common.teleoperators.so100_leader import SO100Leader, SO100LeaderConfig
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.utils import hw_to_dataset_features
from lerobot.robots.lekiwi.config_lekiwi import LeKiwiClientConfig
from lerobot.robots.lekiwi.lekiwi_client import LeKiwiClient
from lerobot.teleoperators.keyboard import KeyboardTeleop, KeyboardTeleopConfig
from lerobot.teleoperators.so100_leader import SO100Leader, SO100LeaderConfig
NB_CYCLES_CLIENT_CONNECTION = 250

8
examples/lekiwi/replay.py
View File

@@ -1,9 +1,9 @@
import time
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.robots.lekiwi.config_lekiwi import LeKiwiClientConfig
from lerobot.common.robots.lekiwi.lekiwi_client import LeKiwiClient
from lerobot.common.utils.robot_utils import busy_wait
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.robots.lekiwi.config_lekiwi import LeKiwiClientConfig
from lerobot.robots.lekiwi.lekiwi_client import LeKiwiClient
from lerobot.utils.robot_utils import busy_wait
robot_config = LeKiwiClientConfig(remote_ip="172.18.134.136", id="lekiwi")
robot = LeKiwiClient(robot_config)

6
examples/lekiwi/teleoperate.py
View File

@@ -1,6 +1,6 @@
from lerobot.common.robots.lekiwi import LeKiwiClient, LeKiwiClientConfig
from lerobot.common.teleoperators.keyboard.teleop_keyboard import KeyboardTeleop, KeyboardTeleopConfig
from lerobot.common.teleoperators.so100_leader import SO100Leader, SO100LeaderConfig
from lerobot.robots.lekiwi import LeKiwiClient, LeKiwiClientConfig
from lerobot.teleoperators.keyboard.teleop_keyboard import KeyboardTeleop, KeyboardTeleopConfig
from lerobot.teleoperators.so100_leader import SO100Leader, SO100LeaderConfig
robot_config = LeKiwiClientConfig(remote_ip="172.18.134.136", id="my_lekiwi")

45
lerobot/common/transport/services_pb2.py
View File

@@ -1,45 +0,0 @@
# Generated by the protocol buffer compiler. DO NOT EDIT!
# NO CHECKED-IN PROTOBUF GENCODE
# source: lerobot/common/transport/services.proto
# Protobuf Python Version: 5.29.0
"""Generated protocol buffer code."""
from google.protobuf import descriptor as _descriptor
from google.protobuf import descriptor_pool as _descriptor_pool
from google.protobuf import runtime_version as _runtime_version
from google.protobuf import symbol_database as _symbol_database
from google.protobuf.internal import builder as _builder
_runtime_version.ValidateProtobufRuntimeVersion(
_runtime_version.Domain.PUBLIC,
5,
29,
0,
'',
'lerobot/common/transport/services.proto'
)
# @@protoc_insertion_point(imports)
_sym_db = _symbol_database.Default()
DESCRIPTOR = _descriptor_pool.Default().AddSerializedFile(b'\n\'lerobot/common/transport/services.proto\x12\ttransport\"L\n\nTransition\x12\x30\n\x0etransfer_state\x18\x01 \x01(\x0e\x32\x18.transport.TransferState\x12\x0c\n\x04\x64\x61ta\x18\x02 \x01(\x0c\"L\n\nParameters\x12\x30\n\x0etransfer_state\x18\x01 \x01(\x0e\x32\x18.transport.TransferState\x12\x0c\n\x04\x64\x61ta\x18\x02 \x01(\x0c\"T\n\x12InteractionMessage\x12\x30\n\x0etransfer_state\x18\x01 \x01(\x0e\x32\x18.transport.TransferState\x12\x0c\n\x04\x64\x61ta\x18\x02 \x01(\x0c\"\x07\n\x05\x45mpty*`\n\rTransferState\x12\x14\n\x10TRANSFER_UNKNOWN\x10\x00\x12\x12\n\x0eTRANSFER_BEGIN\x10\x01\x12\x13\n\x0fTRANSFER_MIDDLE\x10\x02\x12\x10\n\x0cTRANSFER_END\x10\x03\x32\x81\x02\n\x0eLearnerService\x12=\n\x10StreamParameters\x12\x10.transport.Empty\x1a\x15.transport.Parameters0\x01\x12<\n\x0fSendTransitions\x12\x15.transport.Transition\x1a\x10.transport.Empty(\x01\x12\x45\n\x10SendInteractions\x12\x1d.transport.InteractionMessage\x1a\x10.transport.Empty(\x01\x12+\n\x05Ready\x12\x10.transport.Empty\x1a\x10.transport.Emptyb\x06proto3')
_globals = globals()
_builder.BuildMessageAndEnumDescriptors(DESCRIPTOR, _globals)
_builder.BuildTopDescriptorsAndMessages(DESCRIPTOR, 'lerobot.common.transport.services_pb2', _globals)
if not _descriptor._USE_C_DESCRIPTORS:
DESCRIPTOR._loaded_options = None
_globals['_TRANSFERSTATE']._serialized_start=305
_globals['_TRANSFERSTATE']._serialized_end=401
_globals['_TRANSITION']._serialized_start=54
_globals['_TRANSITION']._serialized_end=130
_globals['_PARAMETERS']._serialized_start=132
_globals['_PARAMETERS']._serialized_end=208
_globals['_INTERACTIONMESSAGE']._serialized_start=210
_globals['_INTERACTIONMESSAGE']._serialized_end=294
_globals['_EMPTY']._serialized_start=296
_globals['_EMPTY']._serialized_end=303
_globals['_LEARNERSERVICE']._serialized_start=404
_globals['_LEARNERSERVICE']._serialized_end=661
# @@protoc_insertion_point(module_scope)

8
pyproject.toml
View File

@@ -123,10 +123,10 @@ select = ["E4", "E7", "E9", "F", "I", "N", "B", "C4", "SIM"]
exclude_dirs = [
"tests",
"benchmarks",
"lerobot/common/datasets/push_dataset_to_hub",
"lerobot/common/datasets/v2/convert_dataset_v1_to_v2",
"lerobot/common/policies/pi0/conversion_scripts",
"lerobot/scripts/push_dataset_to_hub.py",
"src/lerobot/datasets/push_dataset_to_hub",
"src/lerobot/datasets/v2/convert_dataset_v1_to_v2",
"src/lerobot/policies/pi0/conversion_scripts",
"src/lerobot/scripts/push_dataset_to_hub.py",
]
skips = ["B101", "B311", "B404", "B603"]

8
lerobot/__init__.py → src/lerobot/__init__.py
View File

@@ -167,10 +167,10 @@ available_datasets = sorted(
set(itertools.chain(*available_datasets_per_env.values(), available_real_world_datasets))
)
# lists all available policies from `lerobot/common/policies`
# lists all available policies from `lerobot/policies`
available_policies = ["act", "diffusion", "tdmpc", "vqbet"]
# lists all available robots from `lerobot/common/robot_devices/robots`
# lists all available robots from `lerobot/robot_devices/robots`
available_robots = [
"koch",
"koch_bimanual",
@@ -179,13 +179,13 @@ available_robots = [
"so101",
]
# lists all available cameras from `lerobot/common/robot_devices/cameras`
# lists all available cameras from `lerobot/robot_devices/cameras`
available_cameras = [
"opencv",
"intelrealsense",
]
# lists all available motors from `lerobot/common/robot_devices/motors`
# lists all available motors from `lerobot/robot_devices/motors`
available_motors = [
"dynamixel",
"feetech",

0
lerobot/__version__.py → src/lerobot/__version__.py
View File

10
lerobot/calibrate.py → src/lerobot/calibrate.py
View File

@@ -31,9 +31,9 @@ from pprint import pformat
import draccus
from lerobot.common.cameras.opencv.configuration_opencv import OpenCVCameraConfig # noqa: F401
from lerobot.common.cameras.realsense.configuration_realsense import RealSenseCameraConfig # noqa: F401
from lerobot.common.robots import ( # noqa: F401
from lerobot.cameras.opencv.configuration_opencv import OpenCVCameraConfig # noqa: F401
from lerobot.cameras.realsense.configuration_realsense import RealSenseCameraConfig # noqa: F401
from lerobot.robots import ( # noqa: F401
Robot,
RobotConfig,
koch_follower,
@@ -42,7 +42,7 @@ from lerobot.common.robots import ( # noqa: F401
so100_follower,
so101_follower,
)
from lerobot.common.teleoperators import ( # noqa: F401
from lerobot.teleoperators import ( # noqa: F401
Teleoperator,
TeleoperatorConfig,
koch_leader,
@@ -50,7 +50,7 @@ from lerobot.common.teleoperators import ( # noqa: F401
so100_leader,
so101_leader,
)
from lerobot.common.utils.utils import init_logging
from lerobot.utils.utils import init_logging
@dataclass

0
lerobot/common/cameras/__init__.py → src/lerobot/cameras/__init__.py
View File

0
lerobot/common/cameras/camera.py → src/lerobot/cameras/camera.py
View File

0
lerobot/common/cameras/configs.py → src/lerobot/cameras/configs.py
View File

0
lerobot/common/cameras/opencv/__init__.py → src/lerobot/cameras/opencv/__init__.py
View File

6
lerobot/common/cameras/opencv/camera_opencv.py → src/lerobot/cameras/opencv/camera_opencv.py
View File

@@ -27,7 +27,7 @@ from typing import Any, Dict, List
import cv2
import numpy as np
from lerobot.common.errors import DeviceAlreadyConnectedError, DeviceNotConnectedError
from lerobot.errors import DeviceAlreadyConnectedError, DeviceNotConnectedError
from ..camera import Camera
from ..utils import get_cv2_backend, get_cv2_rotation
@@ -64,8 +64,8 @@ class OpenCVCamera(Camera):
Example:
```python
from lerobot.common.cameras.opencv import OpenCVCamera
from lerobot.common.cameras.configuration_opencv import OpenCVCameraConfig, ColorMode, Cv2Rotation
from lerobot.cameras.opencv import OpenCVCamera
from lerobot.cameras.configuration_opencv import OpenCVCameraConfig, ColorMode, Cv2Rotation
# Basic usage with camera index 0
config = OpenCVCameraConfig(index_or_path=0)

0
lerobot/common/cameras/opencv/configuration_opencv.py → src/lerobot/cameras/opencv/configuration_opencv.py
View File

0
lerobot/common/cameras/realsense/__init__.py → src/lerobot/cameras/realsense/__init__.py
View File

6
lerobot/common/cameras/realsense/camera_realsense.py → src/lerobot/cameras/realsense/camera_realsense.py
View File

@@ -29,7 +29,7 @@ try:
except Exception as e:
logging.info(f"Could not import realsense: {e}")
from lerobot.common.errors import DeviceAlreadyConnectedError, DeviceNotConnectedError
from lerobot.errors import DeviceAlreadyConnectedError, DeviceNotConnectedError
from ..camera import Camera
from ..configs import ColorMode
@@ -63,8 +63,8 @@ class RealSenseCamera(Camera):
Example:
```python
from lerobot.common.cameras.realsense import RealSenseCamera, RealSenseCameraConfig
from lerobot.common.cameras import ColorMode, Cv2Rotation
from lerobot.cameras.realsense import RealSenseCamera, RealSenseCameraConfig
from lerobot.cameras import ColorMode, Cv2Rotation
# Basic usage with serial number
config = RealSenseCameraConfig(serial_number_or_name="0123456789") # Replace with actual SN

0
lerobot/common/cameras/realsense/configuration_realsense.py → src/lerobot/cameras/realsense/configuration_realsense.py
View File

0
lerobot/common/cameras/utils.py → src/lerobot/cameras/utils.py
View File

6
lerobot/configs/default.py → src/lerobot/configs/default.py
View File

@@ -16,11 +16,11 @@
from dataclasses import dataclass, field
from lerobot.common import (
from lerobot import (
policies, # noqa: F401
)
from lerobot.common.datasets.transforms import ImageTransformsConfig
from lerobot.common.datasets.video_utils import get_safe_default_codec
from lerobot.datasets.transforms import ImageTransformsConfig
from lerobot.datasets.video_utils import get_safe_default_codec
@dataclass

2
lerobot/configs/eval.py → src/lerobot/configs/eval.py
View File

@@ -17,7 +17,7 @@ import logging
from dataclasses import dataclass, field
from pathlib import Path
from lerobot.common import envs, policies # noqa: F401
from lerobot import envs, policies # noqa: F401
from lerobot.configs import parser
from lerobot.configs.default import EvalConfig
from lerobot.configs.policies import PreTrainedConfig

2
lerobot/configs/parser.py → src/lerobot/configs/parser.py
View File

@@ -22,7 +22,7 @@ from typing import Sequence
import draccus
from lerobot.common.utils.utils import has_method
from lerobot.utils.utils import has_method
PATH_KEY = "path"
PLUGIN_DISCOVERY_SUFFIX = "discover_packages_path"

8
lerobot/configs/policies.py → src/lerobot/configs/policies.py
View File

@@ -23,11 +23,11 @@ from huggingface_hub import hf_hub_download
from huggingface_hub.constants import CONFIG_NAME
from huggingface_hub.errors import HfHubHTTPError
from lerobot.common.optim.optimizers import OptimizerConfig
from lerobot.common.optim.schedulers import LRSchedulerConfig
from lerobot.common.utils.hub import HubMixin
from lerobot.common.utils.utils import auto_select_torch_device, is_amp_available, is_torch_device_available
from lerobot.configs.types import FeatureType, NormalizationMode, PolicyFeature
from lerobot.optim.optimizers import OptimizerConfig
from lerobot.optim.schedulers import LRSchedulerConfig
from lerobot.utils.hub import HubMixin
from lerobot.utils.utils import auto_select_torch_device, is_amp_available, is_torch_device_available
# Generic variable that is either PreTrainedConfig or a subclass thereof
T = TypeVar("T", bound="PreTrainedConfig")

8
lerobot/configs/train.py → src/lerobot/configs/train.py
View File

@@ -21,13 +21,13 @@ import draccus
from huggingface_hub import hf_hub_download
from huggingface_hub.errors import HfHubHTTPError
from lerobot.common import envs
from lerobot.common.optim import OptimizerConfig
from lerobot.common.optim.schedulers import LRSchedulerConfig
from lerobot.common.utils.hub import HubMixin
from lerobot import envs
from lerobot.configs import parser
from lerobot.configs.default import DatasetConfig, EvalConfig, WandBConfig
from lerobot.configs.policies import PreTrainedConfig
from lerobot.optim import OptimizerConfig
from lerobot.optim.schedulers import LRSchedulerConfig
from lerobot.utils.hub import HubMixin
TRAIN_CONFIG_NAME = "train_config.json"

0
lerobot/configs/types.py → src/lerobot/configs/types.py
View File

0
lerobot/common/constants.py → src/lerobot/constants.py
View File

4
lerobot/common/datasets/backward_compatibility.py → src/lerobot/datasets/backward_compatibility.py
View File

@@ -20,7 +20,7 @@ The dataset you requested ({repo_id}) is in {version} format.
We introduced a new format since v2.0 which is not backward compatible with v1.x.
Please, use our conversion script. Modify the following command with your own task description:
```
python lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py \\
python -m lerobot.datasets.v2.convert_dataset_v1_to_v2 \\
--repo-id {repo_id} \\
--single-task "TASK DESCRIPTION." # <---- /!\\ Replace TASK DESCRIPTION /!\\
```
@@ -40,7 +40,7 @@ The dataset you requested ({repo_id}) is in {version} format.
While current version of LeRobot is backward-compatible with it, the version of your dataset still uses global
stats instead of per-episode stats. Update your dataset stats to the new format using this command:
```
python lerobot/common/datasets/v21/convert_dataset_v20_to_v21.py --repo-id={repo_id}
python -m lerobot.datasets.v21.convert_dataset_v20_to_v21 --repo-id={repo_id}
```
If you encounter a problem, contact LeRobot maintainers on [Discord](https://discord.com/invite/s3KuuzsPFb)

0
lerobot/common/datasets/card_template.md → src/lerobot/datasets/card_template.md
View File

2
lerobot/common/datasets/compute_stats.py → src/lerobot/datasets/compute_stats.py
View File

@@ -15,7 +15,7 @@
# limitations under the License.
import numpy as np
from lerobot.common.datasets.utils import load_image_as_numpy
from lerobot.datasets.utils import load_image_as_numpy
def estimate_num_samples(

8
lerobot/common/datasets/factory.py → src/lerobot/datasets/factory.py
View File

@@ -18,14 +18,14 @@ from pprint import pformat
import torch
from lerobot.common.datasets.lerobot_dataset import (
from lerobot.configs.policies import PreTrainedConfig
from lerobot.configs.train import TrainPipelineConfig
from lerobot.datasets.lerobot_dataset import (
LeRobotDataset,
LeRobotDatasetMetadata,
MultiLeRobotDataset,
)
from lerobot.common.datasets.transforms import ImageTransforms
from lerobot.configs.policies import PreTrainedConfig
from lerobot.configs.train import TrainPipelineConfig
from lerobot.datasets.transforms import ImageTransforms
IMAGENET_STATS = {
"mean": [[[0.485]], [[0.456]], [[0.406]]], # (c,1,1)

0
lerobot/common/datasets/image_writer.py → src/lerobot/datasets/image_writer.py
View File

12
lerobot/common/datasets/lerobot_dataset.py → src/lerobot/datasets/lerobot_dataset.py
View File

@@ -30,10 +30,10 @@ from huggingface_hub import HfApi, snapshot_download
from huggingface_hub.constants import REPOCARD_NAME
from huggingface_hub.errors import RevisionNotFoundError
from lerobot.common.constants import HF_LEROBOT_HOME
from lerobot.common.datasets.compute_stats import aggregate_stats, compute_episode_stats
from lerobot.common.datasets.image_writer import AsyncImageWriter, write_image
from lerobot.common.datasets.utils import (
from lerobot.constants import HF_LEROBOT_HOME
from lerobot.datasets.compute_stats import aggregate_stats, compute_episode_stats
from lerobot.datasets.image_writer import AsyncImageWriter, write_image
from lerobot.datasets.utils import (
DEFAULT_FEATURES,
DEFAULT_IMAGE_PATH,
INFO_PATH,
@@ -65,7 +65,7 @@ from lerobot.common.datasets.utils import (
write_info,
write_json,
)
from lerobot.common.datasets.video_utils import (
from lerobot.datasets.video_utils import (
VideoFrame,
decode_video_frames,
encode_video_frames,
@@ -357,7 +357,7 @@ class LeRobotDataset(torch.utils.data.Dataset):
the dataset from that address and load it, pending your dataset is compliant with
codebase_version v2.0. If your dataset has been created before this new format, you will be
prompted to convert it using our conversion script from v1.6 to v2.0, which you can find at
lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py.
lerobot/datasets/v2/convert_dataset_v1_to_v2.py.
2. Your dataset doesn't already exists (either on local disk or on the Hub): you can create an empty

2
lerobot/common/datasets/online_buffer.py → src/lerobot/datasets/online_buffer.py
View File

@@ -28,7 +28,7 @@ from typing import Any
import numpy as np
import torch
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.lerobot_dataset import LeRobotDataset
def _make_memmap_safe(**kwargs) -> np.memmap:

2
lerobot/common/datasets/push_dataset_to_hub/utils.py → src/lerobot/datasets/push_dataset_to_hub/utils.py
View File

@@ -23,7 +23,7 @@ import numpy
import PIL
import torch
from lerobot.common.datasets.video_utils import encode_video_frames
from lerobot.datasets.video_utils import encode_video_frames
def concatenate_episodes(ep_dicts):

0
lerobot/common/datasets/sampler.py → src/lerobot/datasets/sampler.py
View File

0
lerobot/common/datasets/transforms.py → src/lerobot/datasets/transforms.py
View File

12
lerobot/common/datasets/utils.py → src/lerobot/datasets/utils.py
View File

@@ -35,14 +35,14 @@ from huggingface_hub.errors import RevisionNotFoundError
from PIL import Image as PILImage
from torchvision import transforms
from lerobot.common.datasets.backward_compatibility import (
from lerobot.configs.types import DictLike, FeatureType, PolicyFeature
from lerobot.datasets.backward_compatibility import (
V21_MESSAGE,
BackwardCompatibilityError,
ForwardCompatibilityError,
)
from lerobot.common.robots import Robot
from lerobot.common.utils.utils import is_valid_numpy_dtype_string
from lerobot.configs.types import DictLike, FeatureType, PolicyFeature
from lerobot.robots import Robot
from lerobot.utils.utils import is_valid_numpy_dtype_string
DEFAULT_CHUNK_SIZE = 1000 # Max number of episodes per chunk
@@ -664,7 +664,7 @@ def create_lerobot_dataset_card(
**kwargs,
) -> DatasetCard:
"""
Keyword arguments will be used to replace values in ./lerobot/common/datasets/card_template.md.
Keyword arguments will be used to replace values in src/lerobot/datasets/card_template.md.
Note: If specified, license must be one of https://huggingface.co/docs/hub/repositories-licenses.
"""
card_tags = ["LeRobot"]
@@ -687,7 +687,7 @@ def create_lerobot_dataset_card(
],
)
card_template = (importlib.resources.files("lerobot.common.datasets") / "card_template.md").read_text()
card_template = (importlib.resources.files("lerobot.datasets") / "card_template.md").read_text()
return DatasetCard.from_template(
card_data=card_data,

4
lerobot/common/datasets/v2/batch_convert_dataset_v1_to_v2.py → src/lerobot/datasets/v2/batch_convert_dataset_v1_to_v2.py
View File

@@ -26,8 +26,8 @@ from pathlib import Path
from textwrap import dedent
from lerobot import available_datasets
from lerobot.common.datasets.v2.convert_dataset_v1_to_v2 import convert_dataset
from lerobot.common.robots.aloha.configuration_aloha import AlohaRobotConfig
from lerobot.datasets.v2.convert_dataset_v1_to_v2 import convert_dataset
from lerobot.robots.aloha.configuration_aloha import AlohaRobotConfig
LOCAL_DIR = Path("data/")

22
lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py → src/lerobot/datasets/v2/convert_dataset_v1_to_v2.py
View File

@@ -38,7 +38,7 @@ If your dataset contains a single task, you can simply provide it directly via t
Examples:
```bash
python lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py \
python -m lerobot.datasets.v2.convert_dataset_v1_to_v2 \
--repo-id lerobot/aloha_sim_insertion_human_image \
--single-task "Insert the peg into the socket." \
--robot-config lerobot/configs/robot/aloha.yaml \
@@ -46,7 +46,7 @@ python lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py \
```
```bash
python lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py \
python -m lerobot.datasets.v2.convert_dataset_v1_to_v2 \
--repo-id aliberts/koch_tutorial \
--single-task "Pick the Lego block and drop it in the box on the right." \
--robot-config lerobot/configs/robot/koch.yaml \
@@ -63,7 +63,7 @@ If your dataset is a multi-task dataset, you have two options to provide the tas
Example:
```bash
python lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py \
python -m lerobot.datasets.v2.convert_dataset_v1_to_v2 \
--repo-id lerobot/stanford_kuka_multimodal_dataset \
--tasks-col "language_instruction" \
--local-dir data
@@ -92,7 +92,7 @@ parquet file, and you must provide this column's name with the '--tasks-col' arg
Example:
```bash
python lerobot/common/datasets/v2/convert_dataset_v1_to_v2.py \
python -m lerobot.datasets.v2.convert_dataset_v1_to_v2 \
--repo-id lerobot/stanford_kuka_multimodal_dataset \
--tasks-col "language_instruction" \
--local-dir data
@@ -119,7 +119,7 @@ from huggingface_hub import HfApi
from huggingface_hub.errors import EntryNotFoundError, HfHubHTTPError
from safetensors.torch import load_file
from lerobot.common.datasets.utils import (
from lerobot.datasets.utils import (
DEFAULT_CHUNK_SIZE,
DEFAULT_PARQUET_PATH,
DEFAULT_VIDEO_PATH,
@@ -136,12 +136,12 @@ from lerobot.common.datasets.utils import (
write_json,
write_jsonlines,
)
from lerobot.common.datasets.video_utils import (
from lerobot.datasets.video_utils import (
VideoFrame, # noqa: F401
get_image_pixel_channels,
get_video_info,
)
from lerobot.common.robots import RobotConfig
from lerobot.robots import RobotConfig
V16 = "v1.6"
V20 = "v2.0"
@@ -602,19 +602,19 @@ def make_robot_config(robot_type: str, **kwargs) -> RobotConfig:
raise NotImplementedError # TODO
elif robot_type == "koch_follower":
from lerobot.common.robots.koch_follower import KochFollowerConfig
from lerobot.robots.koch_follower import KochFollowerConfig
return KochFollowerConfig(**kwargs)
elif robot_type == "so100_follower":
from lerobot.common.robots.so100_follower import SO100FollowerConfig
from lerobot.robots.so100_follower import SO100FollowerConfig
return SO100FollowerConfig(**kwargs)
elif robot_type == "stretch":
from lerobot.common.robots.stretch3 import Stretch3RobotConfig
from lerobot.robots.stretch3 import Stretch3RobotConfig
return Stretch3RobotConfig(**kwargs)
elif robot_type == "lekiwi":
from lerobot.common.robots.lekiwi import LeKiwiConfig
from lerobot.robots.lekiwi import LeKiwiConfig
return LeKiwiConfig(**kwargs)
else:

6
lerobot/common/datasets/v21/_remove_language_instruction.py → src/lerobot/datasets/v21/_remove_language_instruction.py
View File

@@ -20,9 +20,9 @@ from datasets import get_dataset_config_info
from huggingface_hub import HfApi
from lerobot import available_datasets
from lerobot.common.datasets.lerobot_dataset import LeRobotDatasetMetadata
from lerobot.common.datasets.utils import INFO_PATH, write_info
from lerobot.common.datasets.v21.convert_dataset_v20_to_v21 import V20, SuppressWarnings
from lerobot.datasets.lerobot_dataset import LeRobotDatasetMetadata
from lerobot.datasets.utils import INFO_PATH, write_info
from lerobot.datasets.v21.convert_dataset_v20_to_v21 import V20, SuppressWarnings
LOCAL_DIR = Path("data/")

2
lerobot/common/datasets/v21/batch_convert_dataset_v20_to_v21.py → src/lerobot/datasets/v21/batch_convert_dataset_v20_to_v21.py
View File

@@ -24,7 +24,7 @@ from pathlib import Path
from huggingface_hub import HfApi
from lerobot import available_datasets
from lerobot.common.datasets.v21.convert_dataset_v20_to_v21 import V21, convert_dataset
from lerobot.datasets.v21.convert_dataset_v20_to_v21 import V21, convert_dataset
LOCAL_DIR = Path("data/")

8
lerobot/common/datasets/v21/convert_dataset_v20_to_v21.py → src/lerobot/datasets/v21/convert_dataset_v20_to_v21.py
View File

@@ -25,7 +25,7 @@ This script will help you convert any LeRobot dataset already pushed to the hub
Usage:
```bash
python lerobot/common/datasets/v21/convert_dataset_v20_to_v21.py \
python -m lerobot.datasets.v21.convert_dataset_v20_to_v21 \
--repo-id=aliberts/koch_tutorial
```
@@ -36,9 +36,9 @@ import logging
from huggingface_hub import HfApi
from lerobot.common.datasets.lerobot_dataset import CODEBASE_VERSION, LeRobotDataset
from lerobot.common.datasets.utils import EPISODES_STATS_PATH, STATS_PATH, load_stats, write_info
from lerobot.common.datasets.v21.convert_stats import check_aggregate_stats, convert_stats
from lerobot.datasets.lerobot_dataset import CODEBASE_VERSION, LeRobotDataset
from lerobot.datasets.utils import EPISODES_STATS_PATH, STATS_PATH, load_stats, write_info
from lerobot.datasets.v21.convert_stats import check_aggregate_stats, convert_stats
V20 = "v2.0"
V21 = "v2.1"

6
lerobot/common/datasets/v21/convert_stats.py → src/lerobot/datasets/v21/convert_stats.py
View File

@@ -17,9 +17,9 @@ from concurrent.futures import ThreadPoolExecutor, as_completed
import numpy as np
from tqdm import tqdm
from lerobot.common.datasets.compute_stats import aggregate_stats, get_feature_stats, sample_indices
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.datasets.utils import write_episode_stats
from lerobot.datasets.compute_stats import aggregate_stats, get_feature_stats, sample_indices
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.datasets.utils import write_episode_stats
def sample_episode_video_frames(dataset: LeRobotDataset, episode_index: int, ft_key: str) -> np.ndarray:

0
lerobot/common/datasets/video_utils.py → src/lerobot/datasets/video_utils.py
View File

0
lerobot/common/envs/__init__.py → src/lerobot/envs/__init__.py
View File

6
lerobot/common/envs/configs.py → src/lerobot/envs/configs.py
View File

@@ -18,10 +18,10 @@ from typing import Any, Optional
import draccus
from lerobot.common.constants import ACTION, OBS_ENV_STATE, OBS_IMAGE, OBS_IMAGES, OBS_STATE
from lerobot.common.robots import RobotConfig
from lerobot.common.teleoperators.config import TeleoperatorConfig
from lerobot.configs.types import FeatureType, PolicyFeature
from lerobot.constants import ACTION, OBS_ENV_STATE, OBS_IMAGE, OBS_IMAGES, OBS_STATE
from lerobot.robots import RobotConfig
from lerobot.teleoperators.config import TeleoperatorConfig
@dataclass

2
lerobot/common/envs/factory.py → src/lerobot/envs/factory.py
View File

@@ -17,7 +17,7 @@ import importlib
import gymnasium as gym
from lerobot.common.envs.configs import AlohaEnv, EnvConfig, HILEnvConfig, PushtEnv, XarmEnv
from lerobot.envs.configs import AlohaEnv, EnvConfig, HILEnvConfig, PushtEnv, XarmEnv
def make_env_config(env_type: str, **kwargs) -> EnvConfig:

4
lerobot/common/envs/utils.py → src/lerobot/envs/utils.py
View File

@@ -22,9 +22,9 @@ import numpy as np
import torch
from torch import Tensor
from lerobot.common.envs.configs import EnvConfig
from lerobot.common.utils.utils import get_channel_first_image_shape
from lerobot.configs.types import FeatureType, PolicyFeature
from lerobot.envs.configs import EnvConfig
from lerobot.utils.utils import get_channel_first_image_shape
def preprocess_observation(observations: dict[str, np.ndarray]) -> dict[str, Tensor]:

0
lerobot/common/errors.py → src/lerobot/errors.py
View File

10
lerobot/find_cameras.py → src/lerobot/find_cameras.py
View File

@@ -37,11 +37,11 @@ from typing import Any, Dict, List
import numpy as np
from PIL import Image
from lerobot.common.cameras.configs import ColorMode
from lerobot.common.cameras.opencv.camera_opencv import OpenCVCamera
from lerobot.common.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.common.cameras.realsense.camera_realsense import RealSenseCamera
from lerobot.common.cameras.realsense.configuration_realsense import RealSenseCameraConfig
from lerobot.cameras.configs import ColorMode
from lerobot.cameras.opencv.camera_opencv import OpenCVCamera
from lerobot.cameras.opencv.configuration_opencv import OpenCVCameraConfig
from lerobot.cameras.realsense.camera_realsense import RealSenseCamera
from lerobot.cameras.realsense.configuration_realsense import RealSenseCameraConfig
logger = logging.getLogger(__name__)

0
lerobot/find_port.py → src/lerobot/find_port.py
View File

0
lerobot/common/model/kinematics.py → src/lerobot/model/kinematics.py
View File

0
lerobot/common/motors/__init__.py → src/lerobot/motors/__init__.py
View File

0
lerobot/common/motors/dynamixel/__init__.py → src/lerobot/motors/dynamixel/__init__.py
View File

2
lerobot/common/motors/dynamixel/dynamixel.py → src/lerobot/motors/dynamixel/dynamixel.py
View File

@@ -22,7 +22,7 @@ import logging
from copy import deepcopy
from enum import Enum
from lerobot.common.utils.encoding_utils import decode_twos_complement, encode_twos_complement
from lerobot.utils.encoding_utils import decode_twos_complement, encode_twos_complement
from ..motors_bus import Motor, MotorCalibration, MotorsBus, NameOrID, Value, get_address
from .tables import (

0
lerobot/common/motors/dynamixel/tables.py → src/lerobot/motors/dynamixel/tables.py
View File

0
lerobot/common/motors/feetech/__init__.py → src/lerobot/motors/feetech/__init__.py
View File

2
lerobot/common/motors/feetech/feetech.py → src/lerobot/motors/feetech/feetech.py
View File

@@ -17,7 +17,7 @@ from copy import deepcopy
from enum import Enum
from pprint import pformat
from lerobot.common.utils.encoding_utils import decode_sign_magnitude, encode_sign_magnitude
from lerobot.utils.encoding_utils import decode_sign_magnitude, encode_sign_magnitude
from ..motors_bus import Motor, MotorCalibration, MotorsBus, NameOrID, Value, get_address
from .tables import (

0
lerobot/common/motors/feetech/tables.py → src/lerobot/motors/feetech/tables.py
View File

6
lerobot/common/motors/motors_bus.py → src/lerobot/motors/motors_bus.py
View File

@@ -32,8 +32,8 @@ import serial
from deepdiff import DeepDiff
from tqdm import tqdm
from lerobot.common.errors import DeviceAlreadyConnectedError, DeviceNotConnectedError
from lerobot.common.utils.utils import enter_pressed, move_cursor_up
from lerobot.errors import DeviceAlreadyConnectedError, DeviceNotConnectedError
from lerobot.utils.utils import enter_pressed, move_cursor_up
NameOrID: TypeAlias = str | int
Value: TypeAlias = int | float
@@ -446,7 +446,7 @@ class MotorsBus(abc.ABC):
except (FileNotFoundError, OSError, serial.SerialException) as e:
raise ConnectionError(
f"\nCould not connect on port '{self.port}'. Make sure you are using the correct port."
"\nTry running `python lerobot/find_port.py`\n"
"\nTry running `python -m lerobot.find_port`\n"
) from e
@abc.abstractmethod

0
lerobot/common/optim/__init__.py → src/lerobot/optim/__init__.py
View File

2
lerobot/common/optim/factory.py → src/lerobot/optim/factory.py
View File

@@ -18,8 +18,8 @@
from torch.optim import Optimizer
from torch.optim.lr_scheduler import LRScheduler
from lerobot.common.policies.pretrained import PreTrainedPolicy
from lerobot.configs.train import TrainPipelineConfig
from lerobot.policies.pretrained import PreTrainedPolicy
def make_optimizer_and_scheduler(

6
lerobot/common/optim/optimizers.py → src/lerobot/optim/optimizers.py
View File

@@ -22,12 +22,12 @@ import draccus
import torch
from safetensors.torch import load_file, save_file
from lerobot.common.constants import (
from lerobot.constants import (
OPTIMIZER_PARAM_GROUPS,
OPTIMIZER_STATE,
)
from lerobot.common.datasets.utils import flatten_dict, unflatten_dict, write_json
from lerobot.common.utils.io_utils import deserialize_json_into_object
from lerobot.datasets.utils import flatten_dict, unflatten_dict, write_json
from lerobot.utils.io_utils import deserialize_json_into_object
@dataclass

6
lerobot/common/optim/schedulers.py → src/lerobot/optim/schedulers.py
View File

@@ -22,9 +22,9 @@ import draccus
from torch.optim import Optimizer
from torch.optim.lr_scheduler import LambdaLR, LRScheduler
from lerobot.common.constants import SCHEDULER_STATE
from lerobot.common.datasets.utils import write_json
from lerobot.common.utils.io_utils import deserialize_json_into_object
from lerobot.constants import SCHEDULER_STATE
from lerobot.datasets.utils import write_json
from lerobot.utils.io_utils import deserialize_json_into_object
@dataclass

0
lerobot/common/policies/__init__.py → src/lerobot/policies/__init__.py
View File

2
lerobot/common/policies/act/configuration_act.py → src/lerobot/policies/act/configuration_act.py
View File

@@ -15,9 +15,9 @@
# limitations under the License.
from dataclasses import dataclass, field
from lerobot.common.optim.optimizers import AdamWConfig
from lerobot.configs.policies import PreTrainedConfig
from lerobot.configs.types import NormalizationMode
from lerobot.optim.optimizers import AdamWConfig
@PreTrainedConfig.register_subclass("act")

8
lerobot/common/policies/act/modeling_act.py → src/lerobot/policies/act/modeling_act.py
View File

@@ -33,10 +33,10 @@ from torch import Tensor, nn
from torchvision.models._utils import IntermediateLayerGetter
from torchvision.ops.misc import FrozenBatchNorm2d
from lerobot.common.constants import ACTION, OBS_IMAGES
from lerobot.common.policies.act.configuration_act import ACTConfig
from lerobot.common.policies.normalize import Normalize, Unnormalize
from lerobot.common.policies.pretrained import PreTrainedPolicy
from lerobot.constants import ACTION, OBS_IMAGES
from lerobot.policies.act.configuration_act import ACTConfig
from lerobot.policies.normalize import Normalize, Unnormalize
from lerobot.policies.pretrained import PreTrainedPolicy
class ACTPolicy(PreTrainedPolicy):

4
lerobot/common/policies/diffusion/configuration_diffusion.py → src/lerobot/policies/diffusion/configuration_diffusion.py
View File

@@ -16,10 +16,10 @@
# limitations under the License.
from dataclasses import dataclass, field
from lerobot.common.optim.optimizers import AdamConfig
from lerobot.common.optim.schedulers import DiffuserSchedulerConfig
from lerobot.configs.policies import PreTrainedConfig
from lerobot.configs.types import NormalizationMode
from lerobot.optim.optimizers import AdamConfig
from lerobot.optim.schedulers import DiffuserSchedulerConfig
@PreTrainedConfig.register_subclass("diffusion")

10
lerobot/common/policies/diffusion/modeling_diffusion.py → src/lerobot/policies/diffusion/modeling_diffusion.py
View File

@@ -33,11 +33,11 @@ from diffusers.schedulers.scheduling_ddim import DDIMScheduler
from diffusers.schedulers.scheduling_ddpm import DDPMScheduler
from torch import Tensor, nn
from lerobot.common.constants import ACTION, OBS_ENV_STATE, OBS_IMAGES, OBS_STATE
from lerobot.common.policies.diffusion.configuration_diffusion import DiffusionConfig
from lerobot.common.policies.normalize import Normalize, Unnormalize
from lerobot.common.policies.pretrained import PreTrainedPolicy
from lerobot.common.policies.utils import (
from lerobot.constants import ACTION, OBS_ENV_STATE, OBS_IMAGES, OBS_STATE
from lerobot.policies.diffusion.configuration_diffusion import DiffusionConfig
from lerobot.policies.normalize import Normalize, Unnormalize
from lerobot.policies.pretrained import PreTrainedPolicy
from lerobot.policies.utils import (
get_device_from_parameters,
get_dtype_from_parameters,
get_output_shape,

46
lerobot/common/policies/factory.py → src/lerobot/policies/factory.py
View File

@@ -18,60 +18,60 @@ import logging
from torch import nn
from lerobot.common.datasets.lerobot_dataset import LeRobotDatasetMetadata
from lerobot.common.datasets.utils import dataset_to_policy_features
from lerobot.common.envs.configs import EnvConfig
from lerobot.common.envs.utils import env_to_policy_features
from lerobot.common.policies.act.configuration_act import ACTConfig
from lerobot.common.policies.diffusion.configuration_diffusion import DiffusionConfig
from lerobot.common.policies.pi0.configuration_pi0 import PI0Config
from lerobot.common.policies.pi0fast.configuration_pi0fast import PI0FASTConfig
from lerobot.common.policies.pretrained import PreTrainedPolicy
from lerobot.common.policies.sac.configuration_sac import SACConfig
from lerobot.common.policies.sac.reward_model.configuration_classifier import RewardClassifierConfig
from lerobot.common.policies.smolvla.configuration_smolvla import SmolVLAConfig
from lerobot.common.policies.tdmpc.configuration_tdmpc import TDMPCConfig
from lerobot.common.policies.vqbet.configuration_vqbet import VQBeTConfig
from lerobot.configs.policies import PreTrainedConfig
from lerobot.configs.types import FeatureType
from lerobot.datasets.lerobot_dataset import LeRobotDatasetMetadata
from lerobot.datasets.utils import dataset_to_policy_features
from lerobot.envs.configs import EnvConfig
from lerobot.envs.utils import env_to_policy_features
from lerobot.policies.act.configuration_act import ACTConfig
from lerobot.policies.diffusion.configuration_diffusion import DiffusionConfig
from lerobot.policies.pi0.configuration_pi0 import PI0Config
from lerobot.policies.pi0fast.configuration_pi0fast import PI0FASTConfig
from lerobot.policies.pretrained import PreTrainedPolicy
from lerobot.policies.sac.configuration_sac import SACConfig
from lerobot.policies.sac.reward_model.configuration_classifier import RewardClassifierConfig
from lerobot.policies.smolvla.configuration_smolvla import SmolVLAConfig
from lerobot.policies.tdmpc.configuration_tdmpc import TDMPCConfig
from lerobot.policies.vqbet.configuration_vqbet import VQBeTConfig
def get_policy_class(name: str) -> PreTrainedPolicy:
"""Get the policy's class and config class given a name (matching the policy class' `name` attribute)."""
if name == "tdmpc":
from lerobot.common.policies.tdmpc.modeling_tdmpc import TDMPCPolicy
from lerobot.policies.tdmpc.modeling_tdmpc import TDMPCPolicy
return TDMPCPolicy
elif name == "diffusion":
from lerobot.common.policies.diffusion.modeling_diffusion import DiffusionPolicy
from lerobot.policies.diffusion.modeling_diffusion import DiffusionPolicy
return DiffusionPolicy
elif name == "act":
from lerobot.common.policies.act.modeling_act import ACTPolicy
from lerobot.policies.act.modeling_act import ACTPolicy
return ACTPolicy
elif name == "vqbet":
from lerobot.common.policies.vqbet.modeling_vqbet import VQBeTPolicy
from lerobot.policies.vqbet.modeling_vqbet import VQBeTPolicy
return VQBeTPolicy
elif name == "pi0":
from lerobot.common.policies.pi0.modeling_pi0 import PI0Policy
from lerobot.policies.pi0.modeling_pi0 import PI0Policy
return PI0Policy
elif name == "pi0fast":
from lerobot.common.policies.pi0fast.modeling_pi0fast import PI0FASTPolicy
from lerobot.policies.pi0fast.modeling_pi0fast import PI0FASTPolicy
return PI0FASTPolicy
elif name == "sac":
from lerobot.common.policies.sac.modeling_sac import SACPolicy
from lerobot.policies.sac.modeling_sac import SACPolicy
return SACPolicy
elif name == "reward_classifier":
from lerobot.common.policies.sac.reward_model.modeling_classifier import Classifier
from lerobot.policies.sac.reward_model.modeling_classifier import Classifier
return Classifier
elif name == "smolvla":
from lerobot.common.policies.smolvla.modeling_smolvla import SmolVLAPolicy
from lerobot.policies.smolvla.modeling_smolvla import SmolVLAPolicy
return SmolVLAPolicy
else:

0
lerobot/common/policies/normalize.py → src/lerobot/policies/normalize.py
View File

8
lerobot/common/policies/pi0/configuration_pi0.py → src/lerobot/policies/pi0/configuration_pi0.py
View File

@@ -14,12 +14,12 @@
from dataclasses import dataclass, field
from lerobot.common.optim.optimizers import AdamWConfig
from lerobot.common.optim.schedulers import (
CosineDecayWithWarmupSchedulerConfig,
)
from lerobot.configs.policies import PreTrainedConfig
from lerobot.configs.types import FeatureType, NormalizationMode, PolicyFeature
from lerobot.optim.optimizers import AdamWConfig
from lerobot.optim.schedulers import (
CosineDecayWithWarmupSchedulerConfig,
)
@PreTrainedConfig.register_subclass("pi0")

4
lerobot/common/policies/pi0/conversion_scripts/benchmark.py → src/lerobot/policies/pi0/conversion_scripts/benchmark.py
View File

@@ -14,9 +14,9 @@
import torch
from lerobot.common.datasets.lerobot_dataset import LeRobotDataset
from lerobot.common.policies.factory import make_policy
from lerobot.configs.policies import PreTrainedConfig
from lerobot.datasets.lerobot_dataset import LeRobotDataset
from lerobot.policies.factory import make_policy
torch.backends.cudnn.benchmark = True

6
lerobot/common/policies/pi0/conversion_scripts/compare_with_jax.py → src/lerobot/policies/pi0/conversion_scripts/compare_with_jax.py
View File

@@ -18,9 +18,9 @@ from pathlib import Path
import torch
from lerobot.common.datasets.lerobot_dataset import LeRobotDatasetMetadata
from lerobot.common.policies.factory import make_policy
from lerobot.configs.policies import PreTrainedConfig
from lerobot.datasets.lerobot_dataset import LeRobotDatasetMetadata
from lerobot.policies.factory import make_policy
def display(tensor: torch.Tensor):
@@ -97,7 +97,7 @@ def main():
noise = torch.from_numpy(noise).to(device=device, dtype=torch.float32)
from lerobot.common import policies # noqa
from lerobot import policies # noqa
cfg = PreTrainedConfig.from_pretrained(ckpt_torch_dir)
cfg.pretrained_path = ckpt_torch_dir

Some files were not shown because too many files have changed in this diff Show More