This message was deleted.

@Brandon Williams ^^

On a tangential note - I saw this project. They claim to run notebooks fast. https://github.com/fastai/execnb It is used by nbprocess for eg. https://github.com/fastai/nbprocess

interesting. i took a look at the source code. papermill relies on a kernel to execute code (which requires a messaging queue, adding a bit of overhead). execnb executes the code in the same process - it eliminates the overhead of the messaging queue but I'm unsure how fast that can make it. nbprocess seems to be package to export notebooks to quarto thanks for sharing!

I haven’t really played with nbprocess (newer version of nbdev), but I do the find the concept interesting. With nbprocess - the actual code is written in python notebooks mixed with exploration and tests - and that notebook is used to emit python module with only exported cells. nbprocess is evolution of nbdev and will become nbdev2 but nbdev page has better motivation doc right now: https://nbdev.fast.ai/ For eg: Notebook with few exported cells: https://github.com/fastai/nbprocess/blob/master/nbs/06_merge.ipynb This results in this python file: https://github.com/fastai/nbprocess/blob/master/nbprocess/merge.py They also use the same notebook to generate documentation too using quarto to render notebooks. In anycase, it is tangential, but looked interesting to share :)

amazing! will be super useful, looking forward to it

@Amardeep Singh: nbdev is an amazing project! many ploomber users also use nbdev. I've run a few examples but looks like I need to take a closer look, especially now that nbdev2 is coming