hope

README.md

@@ -11,3 +11,29 @@ license: mit
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# PaliGemma Image Captioning Gradio App
+
+## Deployment Instructions
+
+1. Create a new Hugging Face Space
+2. Choose Python as the SDK
+3. Select 16GB CPU hardware
+4. Upload the following files:
+   - `app.py`
+   - `requirements.txt`
+
+### HuggingFace Access Token
+
+1. Go to HuggingFace settings
+2. Create a new access token with "Read" permissions
+3. Add the token as a secret named `HF_TOKEN` in your Space settings
+
+### Features
+- Multi-language image captioning
+- Upload custom images
+- Example images included
+- Supports English, Spanish, French, German captions
+
+## Model Details
+- Model: google/paligemma-3b-mix-224
+- Task: Multilingual Image Captioning

app.py

@@ -1,13 +1,17 @@
+import os
 import gradio as gr
 from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
 from PIL import Image
 import torch
 import requests
 
+# Get token from environment variable
+HF_TOKEN = os.getenv('HF_TOKEN')
+
 # Load the model and processor
 model_id = "google/paligemma-3b-mix-224"
-model = PaliGemmaForConditionalGeneration.from_pretrained(model_id, token=True).eval()
-processor = AutoProcessor.from_pretrained(model_id, token=True)
+model = PaliGemmaForConditionalGeneration.from_pretrained(model_id, token=HF_TOKEN).eval()
+processor = AutoProcessor.from_pretrained(model_id, token=HF_TOKEN)
 
 # Supported languages and example prompts
 LANGUAGES = {