> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/xinntao/Real-ESRGAN/llms.txt
> Use this file to discover all available pages before exploring further.

# Training Overview

> Understand the two-stage training process for Real-ESRGAN

Real-ESRGAN uses a two-stage training process to achieve high-quality image super-resolution. This approach combines the stability of L1 loss with the perceptual quality improvements from GAN training.

## Training Process

The training is divided into two distinct stages that share the same data synthesis process and training pipeline, but differ in their loss functions:

<Steps>
  <Step title="Stage 1: Train Real-ESRNet">
    Train Real-ESRNet using L1 loss from a pre-trained ESRGAN model. This stage provides a stable foundation and prevents mode collapse.

    * Uses L1 loss only
    * Starts from pre-trained ESRGAN weights
    * Results in a stable base model
  </Step>

  <Step title="Stage 2: Train Real-ESRGAN">
    Use the trained Real-ESRNet model as generator initialization and train Real-ESRGAN with a combination of losses.

    * Combines L1 loss, perceptual loss, and GAN loss
    * Improves perceptual quality
    * Produces the final Real-ESRGAN model
  </Step>
</Steps>

## Why Two Stages?

This two-stage approach offers several advantages:

* **Stability**: Starting with L1 loss provides a stable baseline before introducing adversarial training
* **Quality**: The combination of losses in stage 2 improves perceptual quality while maintaining fidelity
* **Convergence**: Pre-training with L1 loss helps the GAN training converge more reliably

## Training Requirements

### Hardware

* Multiple GPUs recommended (examples use 4 GPUs)
* Single GPU training is supported but slower
* Adequate disk space for datasets

### Datasets

Real-ESRGAN is trained on:

* **DF2K**: Combination of DIV2K and Flickr2K datasets
* **OST**: OpenImages subset for training

Only high-resolution (HR) images are required. Low-quality images are generated on-the-fly during training using Real-ESRGAN's degradation process.

<Note>
  The degradation process simulates real-world image degradation, including blur, noise, compression artifacts, and downsampling.
</Note>

## Training Modes

### Debug Mode

Test your configuration before full training:

```bash theme={null}
python realesrgan/train.py -opt options/train_realesrnet_x4plus.yml --debug
```

### Full Training

Run the complete training process:

```bash theme={null}
python realesrgan/train.py -opt options/train_realesrnet_x4plus.yml --auto_resume
```

The `--auto_resume` flag automatically resumes training from the last checkpoint if interrupted.

## Next Steps

<CardGroup cols={2}>
  <Card title="Dataset Preparation" icon="database" href="/training/dataset-preparation">
    Learn how to prepare and process training datasets
  </Card>

  <Card title="Train Real-ESRNet" icon="chart-line" href="/training/train-realesrnet">
    Start with stage 1 training
  </Card>

  <Card title="Train Real-ESRGAN" icon="sparkles" href="/training/train-realesrgan">
    Complete stage 2 for the final model
  </Card>

  <Card title="Fine-tuning" icon="sliders" href="/training/finetune">
    Adapt the model to your custom dataset
  </Card>
</CardGroup>