## Why would you want to install some packages in your base conda environment

In a pinch, you might want to muck around on your system with some quick-and-dirty experiment. Having a suite of packages inside your base environment can be handy. It's like having a scratch environment available.

## How to bootstrap your base conda environment

I would recommend bootstrapping your base anaconda environment with some basic data science packages.

conda activate base
conda install -c conda-forge \
scipy numpy pandas matplotlib \
numpy jupyter jupyterlab \
scikit-learn ipython ipykernel \
ipywidgets mamba


Doing so gives you an environment where you can quickly prototype new things without necessarily going through the overhead of creating an entirely new project (and with it, a full conda environment). Of course, the alternative is to set up a scratch environment, in which you install packages on-the-fly.

Installing mamba can be helpful if you want a faster drop-in replacement for conda. (see: Use Mamba as a faster drop-in replacement for conda for more information.)