README.md

---
title: AutoLineDigitizer API
emoji: 📈
colorFrom: blue
colorTo: green
sdk: docker
pinned: false
license: mit
---

# AutoLineDigitizer API

Automatic chart line data extraction using deep learning (LineFormer + ChartDete + EasyOCR).

Upload a chart image to extract line data that can be imported into [StarryDigitizer](https://digitizer.starrydata.org/).

## API Usage

### Endpoint

```
POST https://<your-space>.hf.space/api/predict
```

### Python Example

```python
from gradio_client import Client

client = Client("https://<your-space>.hf.space/")
result = client.predict(
    "chart.png",               # image path
    True,                      # auto_axis_detection
    "arc_length",              # downsample_mode
    20,                        # max_points
    10,                        # fixed_step
    "mean_y_desc",             # sort_mode
    "starry_digitizer_json",   # output_format
    api_name="/predict"
)
```

### JavaScript (fetch) Example

```javascript
const formData = new FormData();
formData.append('data', JSON.stringify([
  null,   // image placeholder
  true,   // auto_axis_detection
  "arc_length",
  20,
  10,
  "mean_y_desc",
  "starry_digitizer_json"
]));
formData.append('files', imageFile);

const res = await fetch('https://<your-space>.hf.space/api/predict', {
  method: 'POST',
  body: formData
});
```

## Output Formats

- **starry_digitizer_json**: StarryDigitizer project.json (for API integration)
- **starry_digitizer_zip**: ZIP file containing image.png + project.json
- **json**: Raw extraction result with line points and axis calibration