Update app.py

app.py

@@ -4,8 +4,11 @@ import torch
 from PIL import Image
 from transformers import MllamaForConditionalGeneration, AutoProcessor
 from huggingface_hub import login
-HF_TOKEN=st.secrets["newfinegrained"]
+
+# Authenticate with Hugging Face
+HF_TOKEN = st.secrets["newfinegrained"]
 login(HF_TOKEN)
+
 def load_model_and_processor(model_id):
     """Load the model and processor."""
     model = MllamaForConditionalGeneration.from_pretrained(
@@ -17,7 +20,7 @@ def load_model_and_processor(model_id):
     return model, processor
 
 def generate_text(model, processor, image_url, prompt):
-     """Generate text using the model and processor."""
+    """Generate text using the model and processor."""
     try:
         image = Image.open(requests.get(image_url, stream=True).raw)
         inputs = processor(image, prompt, return_tensors="pt").to(model.device)
@@ -32,12 +35,18 @@ st.title("LLaMA 3 Vision Haiku Generator")
 # Model ID and loading
 MODEL_ID = "meta-llama/Llama-3.2-11B-Vision"
 model, processor = load_model_and_processor(MODEL_ID)
-print(model)
-# User input for image URL and prompt
-image_url = st.text_input("Enter the Image URL:", "https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg")
-
-prompt = st.text_area("Enter your prompt:", "<|image|><|begin_of_text|>If I had to write a haiku for this one")
 
+# User input for image URL and prompt
+image_url = st.text_input(
+    "Enter the Image URL:",
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg"
+)
+prompt = st.text_area(
+    "Enter your prompt:",
+    "<|image|><|begin_of_text|>If I had to write a haiku for this one"
+)
+
+# Button to generate haiku
 if st.button("Generate Haiku"):
     with st.spinner("Generating haiku..."):
         result = generate_text(model, processor, image_url, prompt)
@@ -53,6 +62,8 @@ if st.button("Generate Haiku"):
 
 
 
+
+