README.md

---
license: apache-2.0
tags:
- pytorch
- swint
- chest-ct
- survival-analysis
- time-to-event
- model-2d
model-index:
- name: swinT
  results: []
---

# SwinT-CT Checkpoint

This is a PyTorch Lightning `.ckpt` checkpoint for a SwinT model trained on chest CT images.

## Usage

A quickstart script is below.

```python
import radfusion3
dm = radfusion3.data.DataModule(config, test_split=config.test_split)
model = radfusion3.builder.build_lightning_model(config, ckpt=ckpt)
trainer.fit(model=model, datamodule=dm, ckpt_path=config.ckpt)
trainer.test(datamodule=dm, ckpt_path="best")
```

For detailed instructions please follow the [README in Github repo](https://github.com/som-shahlab/INSPECT_public/tree/main?tab=readme-ov-file#1-image-modality-experiment).