Ruby and Jupyter notebooks

Today, I was working through Andrew Ng’s excellent short course on building systems with ChatGPT. The course uses Jupyter notebooks for running the Python code.

While figuring out Python syntax, one question came to mind: does Ruby have something similar to Jupyter notebooks? Turns out that you can just use Jupyter notebooks with Ruby if you setup the iruby kernel. This kernel is available as a gem, and the setup is really simple.

Setting up Jupyter and iruby

First, install Jupyter using Python’s pip installer.

pip install jupyter

Next, install the iruby gem.

gem install iruby

Finally, register iruby as a kernel for Jupyter.

iruby register --force

Open the jupyter notebook:

jupyter notebook

Now if you go Kernel -> Change kernel in the toolbar, you will see Ruby as one of the options.

At this point, we’re ready to create our first Ruby notebook. When you create a new notebook, you will see Ruby as one of the available kernels. You can pick that and run puts "Hello, world!" just to make sure you have it working correctly.

Creating our first notebook

I wanted to play around with the OpenAI API, so I need to install the ruby-openai gem. This can be done in the notebook via bundler’s inline gemfile definition. Put this in the first cell:

require "bundler/inline"

gemfile do
  source "https://rubygems.org"

  gem "dotenv", require: "dotenv/load"
  gem "ruby-openai"
end

Aside from ruby-openai, I have also included dotenv here so that I can store OpenAI credentials in a .env file in the notebooks folder. This .env file should look like:

OPENAI_ACCESS_TOKEN=your_key

The require: "dotenv/load" in the gemfile definition reads the .env file and updates the environment with it. We can now use it to initialize the OpenAI client:

client = OpenAI::Client.new(access_token: ENV["OPENAI_ACCESS_TOKEN"])

And use one of the examples from the ruby-openai README:

response = client.chat(
  parameters: {
    model: "gpt-3.5-turbo",
    messages: [{ role: "user", content: "Hello!"}],
    temperature: 0.7,
  }
)

If you have a valid OpenAI access token, you should see a response containing something like “Hello! How can I assist you today?”.

And that’s it! That’s all it took to get to a working playground for exploring ChatGPT. I’m excited to see how I can make notebooks a part of my workflow. I can see notebooks being useful when building up small scripts, or when learning about how a feature of Ruby works. I’m also considering using a notebook to draft this year’s what’s new in Ruby article using markdown cells mixed with code blocks.