### Multiple GPUs Training Setup

Source: https://github.com/cxh-research/docunfold/blob/main/README.md

Configure multi-GPU training using the `accelerate` library. This command initiates the configuration process.

```bash
accelerate config
```

--------------------------------

### Multiple GPUs Training Launch

Source: https://github.com/cxh-research/docunfold/blob/main/README.md

Launch the training process across multiple GPUs after configuration with `accelerate config`. Refer to the Accelerate documentation for detailed usage.

```bash
accelerate launch train.py
```

--------------------------------

### Single GPU Training

Source: https://github.com/cxh-research/docunfold/blob/main/README.md

Use this command for training the model on a single GPU. Ensure the dataset and save directories are configured in `config.yml`.

```bash
python train.py
```

--------------------------------

### Inference

Source: https://github.com/cxh-research/docunfold/blob/main/README.md

Run this command to perform inference with the trained DocUnfold model. Ensure the model is properly set up for inference.

```bash
python infer.py
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.