Update app.py

app.py

@@ -1,53 +1,6 @@
 # Importing the requirements
 import gradio as gr
-import torch
-from PIL import Image
-from transformers import AutoModel, AutoTokenizer
-import spaces
-
-# Device for the model
-device = "cuda"
-
-# Load the model and tokenizer
-model = AutoModel.from_pretrained(
-    "openbmb/MiniCPM-Llama3-V-2_5", trust_remote_code=True, torch_dtype=torch.float16
-)
-model = model.to(device="cuda")
-tokenizer = AutoTokenizer.from_pretrained(
-    "openbmb/MiniCPM-Llama3-V-2_5", trust_remote_code=True
-)
-model.eval()
-
-
-@spaces.GPU(duration=120)
-def answer_question(image, question):
-    """
-    Generates an answer to a given question based on the provided image and question.
-
-    Args:
-        image (str): The path to the image file.
-        question (str): The question text.
-
-    Returns:
-        str: The generated answer to the question.
-    """
-
-    # Message format for the model
-    msgs = [{"role": "user", "content": question}]
-
-    # Generate the answer
-    res = model.chat(
-        image=image,
-        msgs=msgs,
-        tokenizer=tokenizer,
-        sampling=True,
-        temperature=0.7,
-        stream=True,
-        system_prompt="You are an AI assistant specialized in visual content analysis. Given an image and a related question, analyze the image thoroughly and provide a precise and informative answer based on the visible content. Ensure your response is clear, accurate, and directly addresses the question.",
-    )
-
-    # Return the answer
-    return "".join(res)
+from model import answer_question
 
 
 # Image and text inputs for the interface
@@ -59,9 +12,9 @@ answer = gr.Textbox(label="Predicted answer")
 
 # Examples for the interface
 examples = [
-    ["cat.jpg", "How many cats are there?"],
-    ["dog.jpg", "What color is the dog?"],
-    ["bird.jpg", "What is the bird doing?"],
+    ["images/cat.jpg", "How many cats are there?"],
+    ["images/dog.jpg", "What color is the dog?"],
+    ["images/bird.jpg", "What is the bird doing?"],
 ]
 
 # Title, description, and article for the interface