Spaces:
Sleeping
Sleeping
import os | |
from PIL import Image | |
import streamlit as st | |
from streamlit_option_menu import option_menu | |
from gemini_utility import (load_gemini_pro_model, | |
gemini_pro_response, | |
gemini_pro_vision_response, | |
embeddings_model_response) | |
working_dir = os.path.dirname(os.path.abspath(__file__)) | |
st.set_page_config( | |
page_title="Gemini AI", | |
page_icon="π§ ", | |
layout="centered", | |
) | |
with st.sidebar: | |
selected = option_menu('Gemini AI', | |
['ChatBot', | |
'Image Captioning', | |
'Embed text', | |
'Ask me anything'], | |
menu_icon='robot', icons=['chat-dots-fill', 'image-fill', 'textarea-t', 'patch-question-fill'], | |
default_index=0 | |
) | |
# Function to translate roles between Gemini-Pro and Streamlit terminology | |
def translate_role_for_streamlit(user_role): | |
if user_role == "model": | |
return "assistant" | |
else: | |
return user_role | |
# chatbot page | |
if selected == 'ChatBot': | |
model = load_gemini_pro_model() | |
# Initialize chat session in Streamlit if not already present | |
if "chat_session" not in st.session_state: # Renamed for clarity | |
st.session_state.chat_session = model.start_chat(history=[]) | |
# Display the chatbot's title on the page | |
st.title("π€ ChatBot") | |
# Display the chat history | |
for message in st.session_state.chat_session.history: | |
with st.chat_message(translate_role_for_streamlit(message.role)): | |
st.markdown(message.parts[0].text) | |
# Input field for user's message | |
user_prompt = st.chat_input("Ask Gemini-Pro...") # Renamed for clarity | |
if user_prompt: | |
# Add user's message to chat and display it | |
st.chat_message("user").markdown(user_prompt) | |
# Send user's message to Gemini-Pro and get the response | |
gemini_response = st.session_state.chat_session.send_message(user_prompt) # Renamed for clarity | |
# Display Gemini-Pro's response | |
with st.chat_message("assistant"): | |
st.markdown(gemini_response.text) | |
# Image captioning page | |
if selected == "Image Captioning": | |
st.title("π· Snap Narrate") | |
uploaded_image = st.file_uploader("Upload an image...", type=["jpg", "jpeg", "png"]) | |
if st.button("Generate Caption"): | |
image = Image.open(uploaded_image) | |
col1, col2 = st.columns(2) | |
with col1: | |
resized_img = image.resize((800, 500)) | |
st.image(resized_img) | |
default_prompt = "write a short caption for this image" # change this prompt as per your requirement | |
# get the caption of the image from the gemini-pro-vision LLM | |
caption = gemini_pro_vision_response(default_prompt, image) | |
with col2: | |
st.info(caption) | |
# text embedding model | |
if selected == "Embed text": | |
st.title("π‘ Embed Text") | |
# text box to enter prompt | |
user_prompt = st.text_area(label='', placeholder="Enter the text to get embeddings") | |
if st.button("Get Response"): | |
response = embeddings_model_response(user_prompt) | |
st.markdown(response) | |
# text embedding model | |
if selected == "Ask me anything": | |
st.title("β Ask me a question") | |
# text box to enter prompt | |
user_prompt = st.text_area(label='', placeholder="Ask me anything...") | |
if st.button("Get Response"): | |
response = gemini_pro_response(user_prompt) | |
st.markdown(response) |