Frictionless hyperparameter tuning with W&B and Jupyter Notebooks


I’ve just started a blog titled " Machine Learning Tip of The Day", where I’ll irregularly share tips for ML devs that I find while working in my projects. The first post is about Hyperparameter tuning with wandb sweeps and and Jupyter Notebooks!

It is just a small trick to change the program argument of the sweep configuration and replace it with a command that executes the notebook.

1 Like

Thanks for sharing this!

I’m sorry this one slipped through the cracks and didn’t get a response. This is a really cool usage of Jupyter nbconvert --execute with Sweeps.

While I’m here, here’s a nice guide within the docs to use Jupyter Notebooks with Sweeps: Running Sweeps in Jupyter - Documentation

1 Like