This repository contains a simple yet powerful chatbot built with Streamlit, OpenAI, and LangChain. The chatbot maintains conversational memory, meaning it can reference past exchanges in its responses.
The chatbot is a demonstration of integrating OpenAI's GPT model, the LangChain library, and Streamlit for creating interactive web applications. The bot's conversational memory allows it to maintain context during the chat session, leading to a more coherent and engaging user experience. Importantly, this feature-rich chatbot application is implemented in less than 40 lines of code (excluding whitespace and comments)!
- Streamlit: A powerful, fast Python framework used to create the web interface for the chatbot.
- OpenAI's GPT: A state-of-the-art language processing AI model that generates the chatbot's responses.
- LangChain: A wrapper library for the ChatGPT model that helps manage conversation history and structure the model's responses.
- Python 3.6 or higher
- Streamlit
- LangChain
- OpenAI API key
- Clone this repository.
- Install the necessary Python packages using the command
pip install -r requirements.txt. - Set the environment variable for your OpenAI API key.
- Run the Streamlit app using the command
streamlit run streamlit_app.py.
The chatbot starts with a system message that sets the tone for the conversation. It then alternates between receiving user inputs and generating AI responses. The conversation history is stored and used as context for generating future responses, allowing the chatbot to maintain conversational continuity.
Contributions, issues, and feature requests are welcome. Feel free to check the Issues page if you want to contribute.
This project is licensed under the terms of the MIT license.
We provide a sample app using Python that you can deploy on App Platform. These steps will get this sample application running for you using App Platform.
Note: Following these steps may result in charges for the use of DigitalOcean services.
- You need a DigitalOcean account. If you don't already have one, you can sign up at https://cloud.digitalocean.com/registrations/new.
Click this button to deploy the app to the DigitalOcean App Platform. If you are not logged in, you will be prompted to log in with your DigitalOcean account.
Using this button disables the ability to automatically re-deploy your app when pushing to a branch or tag in your repository as you are using this repo directly.
If you want to automatically re-deploy your app, fork the GitHub repository to your account so that you have a copy of it stored to the cloud. Click the Fork button in the GitHub repository and follow the on-screen instructions.
After forking the repo, you should now be viewing this README in your own GitHub org (e.g. https://github.com/<your-org>/sample-python). To deploy the new repo, visit https://cloud.digitalocean.com/apps and click Create App. Then, click GitHub, select the repository you created and select the main branch. App Platform will inspect the code, automatically detect the kind of component to create, and use the correct buildpack to create and deploy a container.
After clicking the Deploy to DigitalOcean button or completing the instructions above to fork the repo, follow these steps:
- Configure the app such as specifying HTTP routes, environment variables or adding a database.
- Provide a name for your app and select which region you want to deploy your app to and click Next. The closest region to you should be selected by default. All App Platform apps are routed through a global CDN. So this will not affect your app performance, unless it needs to talk to external services.
- On the following screen, leave all the fields as they are and click Next.
- Confirm your Plan settings and how many containers you want to launch and click Launch Basic/Pro App.
- You should see a "Building..." progress indicator. You can click View Logs to see more details of the build.
- It can take a few minutes for the build to finish, but you can follow the progress in the Deployments tab.
- Once the build completes successfully, click the Live App link in the header and you should see your running application in a new tab, displaying the home page.
If you followed the steps to fork the repo and used your own copy when deploying the app, you can push changes to your fork and see App Platform automatically re-deploy the update to your app. During these automatic deployments, your application will never pause or stop serving request because App Platform offers zero-downtime deployments.
Here's an example code change you can make for this app:
- Edit
server.pyand replace "Hello!" on line 12 with a different greeting - Commit the change to the
mainbranch. Normally it's a better practice to create a new branch for your change and then merge that branch tomainafter review, but for this demo you can commit to themainbranch directly. - Visit https://cloud.digitalocean.com/apps and navigate to your sample app.
- You should see a "Building..." progress indicator, just like when you first created the app.
- Once the build completes successfully, click the Live App link in the header and you should see your updated application running. You may need to force refresh the page in your browser (e.g. using Shift+Reload).
You can learn more about the App Platform and how to manage and update your application at https://www.digitalocean.com/docs/app-platform/.
When you no longer need this sample application running live, you can delete it by following these steps:
- Visit the Apps control panel at https://cloud.digitalocean.com/apps.
- Navigate to the sample app.
- In the Settings tab, click Destroy.
Note: If you do not delete your app, charges for using DigitalOcean services will continue to accrue.