admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-05-16 17:20:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

Place logging under accelerate and update docs

Browse Source
This commit is contained in:
Pepijn
2025-10-10 11:25:53 +02:00
parent 52751e8e6d
commit 95b6035baa
2 changed files with 37 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/multi_gpu_training.mdx
+22 -18
View File
@@ -16,15 +16,15 @@ Or install it with the LeRobot accelerate extra:
pip install lerobot[accelerate]
```
## Configuration
## Configuration (Optional)
Configure accelerate for your hardware setup by running:
You can optionally configure accelerate for your hardware setup by running:
```bash
accelerate config
```
This interactive setup will ask you questions about your training environment (number of GPUs, mixed precision settings, etc.). For a simple multi-GPU setup on a single machine, you can use these recommended settings:
This interactive setup will ask you questions about your training environment (number of GPUs, mixed precision settings, etc.) and saves the configuration for future use. For a simple multi-GPU setup on a single machine, you can use these recommended settings:
- Compute environment: This machine
- Number of machines: 1
@@ -32,26 +32,15 @@ This interactive setup will ask you questions about your training environment (n
- GPU ids to use: (leave empty to use all)
- Mixed precision: fp16 or bf16 (recommended for faster training)
**Note:** You can skip this step and specify parameters directly in the launch command (see Option 1 below).
## Training with Multiple GPUs
You can launch training in two ways:
### Option 1: Using accelerate config (recommended)
### Option 1: Without config (specify parameters directly)
If you ran `accelerate config`, simply launch with:
```bash
accelerate launch $(which lerobot-train) \
--dataset.repo_id=${HF_USER}/my_dataset \
--policy.type=act \
--output_dir=outputs/train/act_multi_gpu \
--job_name=act_multi_gpu \
--wandb.enable=true
```
### Option 2: Without config (specify parameters directly)
If you prefer not to run `accelerate config`, you can specify all parameters in the command:
You can specify all parameters directly in the command without running `accelerate config`:
```bash
accelerate launch \
@@ -61,6 +50,7 @@ accelerate launch \
$(which lerobot-train) \
--dataset.repo_id=${HF_USER}/my_dataset \
--policy.type=act \
--policy.repo_id=${HF_USER}/my_trained_policy \
--output_dir=outputs/train/act_multi_gpu \
--job_name=act_multi_gpu \
--wandb.enable=true
@@ -71,6 +61,20 @@ accelerate launch \
- `--num_processes=2`: Number of GPUs to use
- `--mixed_precision=fp16`: Use fp16 mixed precision (or `bf16` if supported)
### Option 2: Using accelerate config
If you prefer to save your configuration, run `accelerate config` once and then simply launch with:
```bash
accelerate launch $(which lerobot-train) \
--dataset.repo_id=${HF_USER}/my_dataset \
--policy.type=act \
--policy.repo_id=${HF_USER}/my_trained_policy \
--output_dir=outputs/train/act_multi_gpu \
--job_name=act_multi_gpu \
--wandb.enable=true
```
## How It Works
When you launch training with accelerate: