Update app.py

app.py

@@ -2,41 +2,55 @@
 from dotenv import load_dotenv
 load_dotenv()
 
-#Import required packages and libraries
-import streamlit as st #Streamlit to create web application
+# Import required packages and libraries
+import streamlit as st
 import os
-import google.generativeai as genai #The primary LLM model that resides inside genAI library
+import generativeai as genai
 from PIL import Image
 
+GOOGLE_API_KEY = "AIzaSyBcOP9DmVuhp3tzmv-y5WJioaAGW1PQouQ"
 
-GOOGLE_API_KEY="AIzaSyBcOP9DmVuhp3tzmv-y5WJioaAGW1PQouQ"
-
-#Fetch the API key that is defined inside .env
+# Fetch the API key that is defined inside .env
 genai.configure(api_key=GOOGLE_API_KEY)
 
-#Function that loads Gemini Model and returns the response
+# Function that loads Gemini Model and returns the response
 def get_model_response(query, user_image):
     model = genai.GenerativeModel('gemini-pro-vision')
     if query != "":
         model_response = model.generate_content([query, user_image])
     else:
         model_response = model.generate_content(user_image)
-    return model_response.text
+    return model_response
 
-#Set up Streamlit application to visualize results
+# Set up Streamlit application to visualize results
 st.set_page_config(page_title="Visualize Image-Text")
 st.header("Poulta App Image Analyses")
 
-user_input=st.text_input("What do you want to know about the image ? ",key="user_input")
+# Add logo
+logo_image_url = "https://poulta.com/assets/site-images/poulta-logo.png"
+st.image(logo_image_url, width=200)  # Adjust width as needed
+
+user_input = st.text_input("What do you want to know about the image? ", key="user_input")
 uploaded_file = st.file_uploader("Pick your image ...", type=["jpg", "jpeg", "png"])
 
-image=""   
+image = ""
 if uploaded_file is not None:
     image = Image.open(uploaded_file)
     st.image(image, caption="Uploaded Image.", use_column_width=True)
 
-submit=st.button("See the magic ✨")
+submit = st.button("See the magic ✨")
 if submit:
-    model_response=get_model_response(user_input, image)
+    model_response = get_model_response(user_input, image)
     st.subheader("Results here 👇🏻")
-    st.write(model_response)
+    
+    # Check if response has prompt_feedback
+    if hasattr(model_response, 'prompt_feedback'):
+        print("Prompt Feedback:", model_response.prompt_feedback)
+        
+        if model_response.prompt_feedback == "blocked":
+            st.write("The prompt was blocked. Try a different input.")
+        else:
+            st.write("No candidates were returned.")
+    else:
+        # Display the results if available
+        st.write(model_response.text)