I'm trying to set up a Streamlit application using LangChain with the Google Gemini API. Below are the steps I followed and the code I'm using, but I'm running into issues (e.g., indentation errors or setup problems). Can someone help me verify the setup process and ensure the code is correct? •First create a folder and a subfolder •In cmd enter py -3.10 -m venv myvenv • Then myvenv\Scripts\activate to activate the environment • python -m pip install --upgrade pip • pip install --upgrade --quiet langchain-google-genai pillow • pip install streamlit • pip install python-dotenv • then copy the api key from ai for developers google in cmd run streamlit run appname.py In the .env file GOOGLE_API_KEY='' LANGCHAIN_API_KEY="" LANGCHAIN_PROJECT="any name" The above will be run on the cmd,and the below mentioned part is the code which will connect the LLM to the console with the help of language chain which is a framework in python and JavaScript import streamlit as st from dotenv import load_dotenv from langchain_core.prompts import ChatPromptTemplate from langchain_google_genai import ChatGoogleGenerativeAI from langchain_core.output_parsers import StrOutputParser # Load environment variables load_dotenv() # Instantiate the model llm = ChatGoogleGenerativeAI( model="gemini-1.5-flash", temperature=0, max_tokens=None, timeout=None, max_retries=2, ) # Create the prompt template prompt = ChatPromptTemplate.from_messages([ ("system", "You are a chatbot"), ("human", "Question: {question}") ]) # Streamlit app interface st.title('Langchain Demo With Gemini') input_text = st.text_input("Enter your question here") # Output parser output_parser = StrOutputParser() # Build the chain chain = prompt | llm | output_parser # Handle user input if input_text: st.write(chain.invoke({'question': input_text}))``` Continue reading...
