README.md

---
license: apache-2.0
---

## Run
Set conda env.
```
make env
conda activate sam-vit-h-encoder-torchscript
make setup
```

Load the SAM model and convert image encoder to torchscript.
```
python convert_torchscript.py
```

Check `model.pt` in `model_repository/sam_torchscript_fp32/1`.