I always have trouble with Python environments. To be honest I haven’t really found a setup that I like for it.
Nevertheless, I want to try out some Python in Quarto. So here is a short note on how I went about using conda to do that.
I already had a Python environment that I wanted to use. In the quarto yaml, i.e. the header you can specify the Jupyter kernel you would like to use.
For this post it looks like this:
---
title: "Python in Quarto"
author: "Filip Wästberg"
date: "2023-03-03"
categories: [python]
jupyter: pyds
image: minions.png
---
So I had created an environment based on a conda.yaml file like this:
conda env create -f conda_env.yml
And activate it
conda activate pyds
The actual yaml file looked like this:
name: pyds
channels:
- https://repo.anaconda.com/pkgs/snowflake
- nodefaults
dependencies:
- python=3.8
- pip
- pip:
# Snowpark
- snowflake-snowpark-python[pandas]==0.10.0
# Basics
- pandas==1.4.3
- numpy==1.22.3
# ML
- scikit-learn==1.1.1
- lightgbm==3.2.1
- xgboost==1.5.0
# Visualization
...
However, to be able to use this environment in Quarto I have to register it to jupyter.
I did that like this:
python -m ipykernel install --user --name=pyds
Good luck!