Skip to content

Hustenbonbon/langchain-demo

 
 

Repository files navigation

🦜🔗 LangChain & ChatGPT

This is a repository for our workshop. You'll find the slides (here)[https://slides.com/maxtharr/langchain-intro].

A lot of the code and some of the slides are from Johanns presentation at CodeTalks: [https://slides.com/johann-peterhartmann/autonome-agenten/].

Setup

Step 1: get the source, luke!

The workshop uses git and docker to make it easier to use on all platforms. If You don't use docker for religious reasons you can use a local python virtenv / conda configuration instead.

git clone https://github.com/mayflower/codetalks23
cd codetalks23
cp .env.dist .env

Step 2: Get an OpenAI API key

Please ignore this section if you already use openai, and just insert one of your existing keys to the .env file.

Create an account at platform.openai.com. Use your e-mail-adress or an existing google or microsoft account.

Create a new API key at platform.openai.com/account/api-keys and insert it to the .env file.

Step 3: Get a SerpApi-Key to search the internet

Create an account at SerpApi. Get the key here: serpapi.com/manage-api-key Add it to your .env file.

Step 4: Use docker to start working

docker run -it --rm -p 8888:8888 -v "${PWD}":/home/jovyan/work jupyter/minimal-notebook:latest    
# Go! 

The docker output should show you a url that you are able to open.

### Interested in langchain? 

You can get the free german translation of Mark Watsons Book at [hub.mayflower.de/langchain-buch](http://hub.mayflower.de/langchain-buch).

About

Sourcecode for langchain workshop 2023

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 100.0%