admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-05-17 17:50:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Remove dataset and mode from HilSerlEnvConfig to a GymManipulatorConfig to reduce verbose of configs during training

Browse Source
This commit is contained in:
Michel Aractingi
2025-08-05 10:35:54 +02:00
parent 9effc5214f
commit 0710f3a0f1
3 changed files with 135 additions and 104 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/hilserl.mdx
+100 -70
View File
@@ -56,18 +56,22 @@ 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/envs/configs.py`. The configuration is now organized into focused, nested sub-configs:
The training process begins with proper configuration for the HILSerl environment. The main configuration class is `GymManipulatorConfig` in `lerobot/scripts/rl/gym_manipulator.py`, which contains nested `HILSerlRobotEnvConfig` and `DatasetConfig`. The configuration is organized into focused, nested sub-configs:
<!-- prettier-ignore-start -->
```python
class GymManipulatorConfig:
env: HILSerlRobotEnvConfig # Environment configuration (nested)
dataset: DatasetConfig # Dataset recording/replay configuration (nested)
mode: str | None = None # "record", "replay", or None (for training)
class HILSerlRobotEnvConfig(EnvConfig):
robot: RobotConfig | None = None # Main robot agent (defined in `lerobot/robots`)
teleop: TeleoperatorConfig | None = None # Teleoperator agent, e.g., gamepad or leader arm
processor: HILSerlProcessorConfig # Processing pipeline configuration (nested)
dataset: DatasetConfig # Dataset recording/replay configuration (nested)
name: str = "real_robot" # Environment name
mode: str = None # "record", "replay", or None (for training)
device: str = "cuda" # Compute device
fps: int = 30 # Control frequency
# Nested processor configuration
class HILSerlProcessorConfig:
@@ -81,13 +85,12 @@ class HILSerlProcessorConfig:
# Dataset configuration
class DatasetConfig:
repo_id: str | None = None # LeRobot dataset repository ID
repo_id: str # LeRobot dataset repository ID
dataset_root: str | None = None # Local dataset root (optional)
task: str = "" # Task identifier
num_episodes: int = 10 # Number of episodes for recording
episode: int = 0 # Episode index for replay
push_to_hub: bool = True # Whether to push datasets to Hub
fps: int = 10 # Control frequency
task: str # Task identifier
num_episodes: int # Number of episodes for recording
episode: int # Episode index for replay
push_to_hub: bool # Whether to push datasets to Hub
```
<!-- prettier-ignore-end -->
@@ -141,24 +144,30 @@ With the bounds defined, you can safely collect demonstrations for training. Tra
Create a configuration file for recording demonstrations (or edit an existing one like [env_config_so100.json](https://huggingface.co/datasets/aractingi/lerobot-example-config-files/blob/main/env_config_so100.json)):
1. Set `mode` to `"record"`
2. Specify a unique `repo_id` for your dataset (e.g., "username/task_name")
3. Set `num_episodes` to the number of demonstrations you want to collect
4. Set `processor.image_preprocessing.crop_params_dict` to `{}` initially (we'll determine crops later)
5. Configure `robot`, `cameras`, and other hardware settings
1. Set `mode` to `"record"` at the root level
2. Specify a unique `repo_id` for your dataset in the `dataset` section (e.g., "username/task_name")
3. Set `num_episodes` in the `dataset` section to the number of demonstrations you want to collect
4. Set `env.processor.image_preprocessing.crop_params_dict` to `{}` initially (we'll determine crops later)
5. Configure `env.robot`, `env.teleop`, and other hardware settings in the `env` section
Example configuration section:
```json
"mode": "record",
"dataset": {
"repo_id": "username/pick_lift_cube",
"dataset_root": null,
"task": "pick_and_lift",
"num_episodes": 15,
"episode": 0,
"push_to_hub": true,
"fps": 10
{
"env": {
"type": "gym_manipulator",
"fps": 10,
// ... robot, teleop, processor configs ...
},
"dataset": {
"repo_id": "username/pick_lift_cube",
"dataset_root": null,
"task": "pick_and_lift",
"num_episodes": 15,
"episode": 0,
"push_to_hub": true
},
"mode": "record"
}
```
@@ -205,12 +214,16 @@ The gamepad provides a very convenient way to control the robot and the episode
To setup the gamepad, you need to set the `control_mode` to `"gamepad"` and define the `teleop` section in the configuration file.
```json
"teleop": {
"type": "gamepad",
"use_gripper": true
},
"processor": {
"control_mode": "gamepad"
{
"env": {
"teleop": {
"type": "gamepad",
"use_gripper": true
},
"processor": {
"control_mode": "gamepad"
}
}
}
```
@@ -233,13 +246,17 @@ The SO101 leader arm has reduced gears that allows it to move and track the foll
To setup the SO101 leader, you need to set the `control_mode` to `"leader"` and define the `teleop` section in the configuration file.
```json
"teleop": {
"type": "so101_leader",
"port": "/dev/tty.usbmodem585A0077921",
"use_degrees": true
},
"processor": {
"control_mode": "leader"
{
"env": {
"teleop": {
"type": "so101_leader",
"port": "/dev/tty.usbmodem585A0077921",
"use_degrees": true
},
"processor": {
"control_mode": "leader"
}
}
}
```
@@ -271,7 +288,7 @@ python -m lerobot.scripts.rl.gym_manipulator --config_path src/lerobot/configs/e
During recording:
1. The robot will reset to the initial position defined in the configuration file `processor.reset.fixed_reset_joint_positions`
1. The robot will reset to the initial position defined in the configuration file `env.processor.reset.fixed_reset_joint_positions`
2. Complete the task successfully
3. The episode ends with a reward of 1 when you press the "success" button
4. If the time limit is reached, or the fail button is pressed, the episode ends with a reward of 0
@@ -330,13 +347,17 @@ observation.images.front: [180, 250, 120, 150]
Add these crop parameters to your training configuration:
```json
"processor": {
"image_preprocessing": {
"crop_params_dict": {
"observation.images.side": [180, 207, 180, 200],
"observation.images.front": [180, 250, 120, 150]
},
"resize_size": [128, 128]
{
"env": {
"processor": {
"image_preprocessing": {
"crop_params_dict": {
"observation.images.side": [180, 207, 180, 200],
"observation.images.front": [180, 250, 120, 150]
},
"resize_size": [128, 128]
}
}
}
}
```
@@ -367,32 +388,35 @@ python -m lerobot.scripts.rl.gym_manipulator --config_path src/lerobot/configs/r
**Key Parameters for Data Collection**
- **mode**: set it to `"record"` to collect a dataset
- **repo_id**: `"hf_username/dataset_name"`, name of the dataset and repo on the hub
- **num_episodes**: Number of episodes to record
- **processor.reset.number_of_steps_after_success**: Number of additional frames to record after a success (reward=1) is detected
- **fps**: Number of frames per second to record
- **push_to_hub**: Whether to push the dataset to the hub
- **mode**: set it to `"record"` to collect a dataset (at root level)
- **dataset.repo_id**: `"hf_username/dataset_name"`, name of the dataset and repo on the hub
- **dataset.num_episodes**: Number of episodes to record
- **env.processor.reset.number_of_steps_after_success**: Number of additional frames to record after a success (reward=1) is detected
- **env.fps**: Number of frames per second to record
- **dataset.push_to_hub**: Whether to push the dataset to the hub
The `processor.reset.number_of_steps_after_success` parameter is crucial as it allows you to collect more positive examples. When a success is detected, the system will continue recording for the specified number of steps while maintaining the reward=1 label. Otherwise, there won't be enough states in the dataset labeled to 1 to train a good classifier.
The `env.processor.reset.number_of_steps_after_success` parameter is crucial as it allows you to collect more positive examples. When a success is detected, the system will continue recording for the specified number of steps while maintaining the reward=1 label. Otherwise, there won't be enough states in the dataset labeled to 1 to train a good classifier.
Example configuration section for data collection:
```json
{
"mode": "record",
"env": {
"type": "gym_manipulator",
"fps": 10,
"processor": {
"reset": {
"number_of_steps_after_success": 15
}
}
},
"dataset": {
"repo_id": "hf_username/dataset_name",
"dataset_root": "data/your_dataset",
"num_episodes": 20,
"push_to_hub": true,
"fps": 10
"push_to_hub": true
},
"processor": {
"reset": {
"number_of_steps_after_success": 15
}
}
"mode": "record"
}
```
@@ -451,13 +475,17 @@ To use your trained reward classifier, configure the `HILSerlRobotEnvConfig` to
<!-- prettier-ignore-start -->
```python
env_config = HILSerlRobotEnvConfig(
processor=HILSerlProcessorConfig(
reward_classifier=RewardClassifierConfig(
pretrained_path="path_to_your_pretrained_trained_model"
)
config = GymManipulatorConfig(
env=HILSerlRobotEnvConfig(
processor=HILSerlProcessorConfig(
reward_classifier=RewardClassifierConfig(
pretrained_path="path_to_your_pretrained_trained_model"
)
),
# Other environment parameters
),
# Other environment parameters
dataset=DatasetConfig(...),
mode=None # For training
)
```
<!-- prettier-ignore-end -->
@@ -466,11 +494,13 @@ or set the argument in the json config file.
```json
{
"processor": {
"reward_classifier": {
"pretrained_path": "path_to_your_pretrained_model",
"success_threshold": 0.7,
"success_reward": 1.0
"env": {
"processor": {
"reward_classifier": {
"pretrained_path": "path_to_your_pretrained_model",
"success_threshold": 0.7,
"success_reward": 1.0
}
}
}
}