Paligemma Rich Captions

🚀 Fine-tuned PaliGemma Model

This is a fine-tuned version of the PaliGemma model on the google/docci dataset. It generates middle-sized captions (200 - 350 characters) and has fewer hallucinations.

🚀 Quick Start

📦 Installation

You can install the necessary library using the following command:

pip install git+https://github.com/huggingface/transformers

💻 Usage Examples

Basic Usage

from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
from PIL import Image
import requests
import torch

model_id = "gokaygokay/paligemma-rich-captions"

url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/car.jpg?download=true"
image = Image.open(requests.get(url, stream=True).raw)

model = PaliGemmaForConditionalGeneration.from_pretrained(model_id).to('cuda').eval()
processor = AutoProcessor.from_pretrained(model_id)

## prefix
prompt = "caption en"
model_inputs = processor(text=prompt, images=image, return_tensors="pt").to('cuda')
input_len = model_inputs["input_ids"].shape[-1]

with torch.inference_mode():
    generation = model.generate(**model_inputs, max_new_tokens=256, do_sample=False)
    generation = generation[0][input_len:]
    decoded = processor.decode(generation, skip_special_tokens=True)
    print(decoded)

📄 License

This project is licensed under the Apache-2.0 license.