README.md

---
license: mit
metrics:
- bleu
- rouge
- meteor
- bertscore
base_model:
- liuhaotian/llava-v1.5-7b
pipeline_tag: visual-question-answering
---

# visual-qa-tem Model Card

## Model details

**base_model**

We finetune our custom data on LLava-v1.5-7b

See on :[liuhaotian/llava-v1.5-7b](https://huggingface.co/liuhaotian/llava-v1.5-7b)

**Paper or resources for more information:**

Our source code publish on : https://github.com/SmartLab-Roy/visual-qa-tem.git

### Download Model
```python
from huggingface_hub import snapshot_download
import os

# Download the model to local directory
model_path = snapshot_download(
    repo_id="LabSmart/visual-qa-tem",
    cache_dir="./models",  # Local cache directory
    resume_download=True
)

print(f"Model downloaded to: {model_path}")
```
### Quick Start

Reference [LLaVA](https://github.com/haotian-liu/LLaVA.git) for environment setup and CLI inference:

```
python -m llava.serve.cli \
    --model-path "model_path from the download output"\
    --image-file "path/to/your/tem_image.jpg" \
    --load-4bit
```