app.py

import os
from huggingface_hub import login
from transformers import BlipProcessor, BlipForConditionalGeneration



# Get Hugging Face Token from environment variable
hf_token = os.getenv('HF_AUTH_TOKEN')
if not hf_token:
    raise ValueError("Hugging Face token is not set in the environment variables.")
login(token=hf_token)

# Load the processor and model
processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")
import gradio as gr
from diffusers import DiffusionPipeline
import torch
import spaces  # Hugging Face Spaces module

# Initialize the model
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-3.5-medium")
st.title("Image Caption Generator")
st.write("Upload an image or provide an image URL to generate its caption.")

# Option for image upload
img_file = st.file_uploader("Choose an image...", type=["png", "jpg", "jpeg"])
if img_file is not None:
    raw_image = Image.open(img_file).convert('RGB')
text = "a photography of"
inputs = processor(raw_image, text, return_tensors="pt",  padding =True, truncation=True, max_length =250)
out = model.generate(**inputs)
caption = processor.decode(out[0], skip_special_tokens=True)

@spaces.GPU(duration=300)
def generate_image(prompt):
    # Move the model to GPU if available
    device = "cuda" if torch.cuda.is_available() else "cpu"
    pipe.to(device)
    image = pipe(prompt).images[0]
    return image

# Create the Gradio interface
iface = gr.Interface(fn=generate_image, 
                     inputs=caption, 
                     outputs=gr.Image(label="Generated Image"),
                     title="Astronaut in a Jungle Model")

# Launch the interface
iface.launch(share=True)