Metadata-Version: 2.4
Name: uniform-vlm
Version: 0.1.2
Summary: Easy one‑liner training + inference tool for the uniform LoRA VLM.
Author-email: Dillon Murphy <ddmurphy@calpoly.edu>, Kailyn Kragas <kkragas@calpoly.edu>, Sofia Bryukhova <sbryukho@calpoly.edu>, Jasmine Cabrera <jkscabrera2003@gmail.com>
Requires-Python: >=3.9
Description-Content-Type: text/markdown
Requires-Dist: pandas>=2.0
Requires-Dist: huggingface_hub>=0.22
Requires-Dist: transformers>=4.39
Requires-Dist: ms-swift

# uniform‑vlm
A lightweight wrapper Python package for **training** and **inference** with NuExtract‑style LoRA adapters.

```bash
pip install uniform-vlm
```

### Inference
```bash
# CLI
uniform-vlm infer images/ --csv preds.csv

# Python
from uniform_vlm.infer import images_to_csv
images_to_csv("images", "preds.csv")
```

### Training (continue fine‑tuning existing adapter by default)
```bash
uniform-vlm train data/train.csv --image-col path --label-col label_json \
             --output-dir output/my_adapter
```

See the Colab walkthrough ➜ <https://colab.research.google.com/drive/1ndRcS9EMcunvrLorQdP97InvCvkEETjJ?usp=sharing>

---
