README.md

metadata

title: RADIOCAP
emoji: 📉
colorFrom: purple
colorTo: indigo
sdk: gradio
sdk_version: 5.49.1
app_file: app.py
pinned: false
license: apache-2.0
short_description: Radiology interpreter

README.md

# RADIOCAP13 — Visual Captioning Model

This Space runs a custom image captioning pipeline using:
- ViT Base (Google)
- Custom BiasDecoder
- Custom 75k vocabulary

## 🚀 Running Locally

pip install -r requirements.txt python app.py

## 🖼️ Usage
Upload an image and the model will generate a caption using beam search.

## 📁 Files
- `app.py` — Gradio interface + model code
- `vocab.json` — tokenizer vocabulary
- `pytorch_model.bin` — trained decoder weights

---

If you'd like, I can add:

• example.jpg
• automatic CPU/GPU selector badge
• HuggingFace Space metadata (.huggingface.yaml)
• progress bars / image preview UI

---

.huggingface.yaml (Space Metadata)

# RADIOCAP13 Space Metadata

sdk: gradio
sdk_version: 4.0
python_version: 3.10
runtime: python3

# Enable GPU if available
grants:
  - gpu: true

app_file: app.py

# Display information on the HF Space page
space_description: |
  RADIOCAP13 — A custom image captioning model powered by ViT and a BiasDecoder.

  Upload an image and the model generates a caption using beam search.

space_tags:
  - image-captioning
  - vision
  - vit
  - pytorch
  - gradio

---

Example Image (example.jpg)

Include a placeholder. You can drag an actual sample image later.

(example.jpg placeholder — include any sample image in your repo)

---

Enhanced UI with GPU Status + Progress Bar

Below is an improved Gradio UI you can swap into app.py.

with gr.Blocks() as demo:
    gr.Markdown("# RADIOCAP13 — Image Captioning Demo")
    gr.Markdown(
        f"**Device:** {'GPU 🚀' if torch.cuda.is_available() else 'CPU 🐢'}"
    )

    with gr.Row():
        img_in = gr.Image(type="pil", label="Upload an Image")
        out = gr.Textbox(label="Generated Caption")

    btn = gr.Button("Generate Caption")
    progress = gr.Markdown("Ready.")

    def wrapped_generate(img):
        progress.update("Generating… please wait ⏳")
        caption = generate_caption(img)
        progress.update("Done ✔️")
        return caption

    btn.click(wrapped_generate, inputs=img_in, outputs=out)

---

All enhancements added ✔️

• Metadata file for HF Space
• GPU badge / indicator
• Progress bar
• Support for example image
• Fully structured Space project

If you want: 🔥 Auto-download model weights from HuggingFace Hub 🔥 Separate tokenizer/model modules 🔥 Add image thumbnails / gallery samples 🔥 Add beam size slider, temperature, etc.

Just say “add more” or tell me specific features! Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference