Update app.py

app.py

@@ -1,50 +1,58 @@
 import streamlit as st
 from PIL import Image
-import easyocr
 import torch
-from transformers import CLIPProcessor, CLIPModel, pipeline
+import easyocr
+from transformers import CLIPProcessor, CLIPModel
 
-# Load CLIP Model & Processor
+# ---- Load CLIP Model ---- #
 @st.cache_resource
 def load_clip_model():
-    model = CLIPModel.from_pretrained("fxmarty/clip-vision-model-tiny")
+    model = CLIPModel.from_pretrained(
+        "fxmarty/clip-vision-model-tiny", 
+        ignore_mismatched_sizes=True  # Fix model size mismatch
+    )
     processor = CLIPProcessor.from_pretrained("fxmarty/clip-vision-model-tiny")
     return model, processor
 
 model, processor = load_clip_model()
 
-# Initialize OCR
+# ---- Load OCR (EasyOCR) ---- #
 @st.cache_resource
 def load_ocr():
     return easyocr.Reader(['en'])
 
 reader = load_ocr()
 
-# Streamlit App
+# ---- Streamlit UI ---- #
+st.set_page_config(page_title="Multimodal AI Assistant", layout="wide")
 st.title("🖼️ Multimodal AI Assistant")
 st.write("Upload an image and ask a question about it!")
 
-# Upload image
-uploaded_file = st.file_uploader("Upload an image", type=["jpg", "png", "jpeg"])
+# ---- Upload Image ---- #
+uploaded_file = st.file_uploader("📤 Upload an image", type=["jpg", "png", "jpeg"])
 
 if uploaded_file is not None:
     # Display Image
     image = Image.open(uploaded_file)
     st.image(image, caption="Uploaded Image", use_column_width=True)
 
-    # Extract text using OCR
-    with st.spinner("Extracting text from image..."):
+    # Extract Text using OCR
+    with st.spinner("🔍 Extracting text from image..."):
         extracted_text = reader.readtext(uploaded_file, detail=0)
-
-    st.write("### 📝 Extracted Text:", extracted_text)
-
-    # User asks a question
+    
+    st.write("### 📝 Extracted Text:")
+    if extracted_text:
+        st.success(extracted_text)
+    else:
+        st.warning("No readable text found in the image.")
+
+    # ---- Ask a Question About the Image ---- #
     user_question = st.text_input("🤖 Ask a question about the image:")
 
     if user_question:
-        with st.spinner("Analyzing image and question..."):
+        with st.spinner("🔍 Analyzing image and generating response..."):
             inputs = processor(text=[user_question], images=image, return_tensors="pt")
             outputs = model.get_image_features(**inputs)
         
         st.write("### 🏆 AI Response:")
-        st.write("CLIP Model Processed the Image! (Further improvements coming soon)")
+        st.write("CLIP Model has processed the image! (Further improvements coming soon)")