Sign inGet started
← Back to all guides

Building a Streamlit app with LangChain and OpenAI in Deepnote

By Filip Žitný

Updated on July 9, 2024

Jinja Error

This guide will walk you through resolving the Jinja2 version conflict and successfully running your Streamlit app in Deepnote.

Issue description

When running Streamlit commands, you might encounter the following error:

ImportError: Pandas requires version '3.0.0' or newer of 'jinja2' (version '2.11.3' currently installed).

Attempts to uninstall and reinstall Jinja2 may result in the following error:

Found existing installation: Jinja2 2.11.3
Not uninstalling jinja2 at /shared-libs/python3.10/py-core/lib/python3.10/site-packages, outside environment /root/venv
Can't uninstall 'Jinja2'. No files were found to uninstall.

Solution

Install the older version of streamlit for compatibility

import streamlit as st # !pip install streamlit==0.84.0
import openai

openai.api_key = st.secrets["openai_api_key"]

def generate_text(prompt):
    response = openai.Completion.create(
        engine="davinci",
        prompt=prompt,
        max_tokens=50
    )
    return response.choices[0].text.strip()

st.title("LangChain and OpenAI Demo")

user_input = st.text_input("Enter a prompt:")
if user_input:
    output = generate_text(user_input)
    st.write(output)

In case you are still having struggles with jinja error verify you installation:

import streamlit as st
import jinja2

print("Streamlit version:", st.__version__)
print("Jinja2 version:", jinja2.__version__)
# Output: Streamlit version: 0.84.0
# Output: Jinja2 version: 2.11.3

Conclusion

By installing a compatible version of Streamlit, you can work around the Jinja2 version conflict and successfully run your Streamlit app with LangChain and OpenAI in Deepnote. This approach ensures that you can leverage Deepnote’s powerful features while maintaining compatibility with essential libraries.

If you encounter further issues, feel free to contact our support for assistance. Happy taking control over the world with AI in Deepnote!

Filip Žitný

Data Scientist

Follow Filip on Twitter, LinkedIn and GitHub

That’s it, time to try Deepnote

Get started – it’s free
Book a demo

Footer

Solutions

  • Notebook
  • Data apps
  • Machine learning
  • Data teams

Product

Company

Comparisons

Resources

  • Privacy
  • Terms

© Deepnote