To configure your environment you will need Anaconda, the Python Distribution.
The instructions for installing Anaconda can be found here
Once Anaconda is installed you should have conda executable in your environment path.
Anaconda provides a concept called environments which allow us to have different dependencies based on what we're working on. The documentation is available here
We would like to use different conda environment for running jobs on our machines, but when we submit jobs to Azure Machine Learning we will use a different environment. To provide easy access to these environments in jupter notebooks or jupyter lab we can use nb_conda into Anaconda's base environment.
conda install -n base nb_conda