Georgetown University Math & Statistics 511

the jupyter notebooks in this repository form the lecture notes for the GU511 course. like any jupyter notebook, they exist in two states:

1. rendered: click on any of the .ipynb files in the repo above and you will be taken to a github-rendered version of the notebook for your own use
2. executable: load any of these .ipynb files into an active jupyter notebook server session

you should consider the Microsoft Azure Notebooks Server to be the default location of these files. If you are having problems with that site, the backup location (where notebooks are rendered but not clonable nor executable) is the github repo.

advanced usage: running these notebooks

Microsoft Azure Notebooks

All of these notebooks are also hosted on Microsoft Azure Notebooks (a preview service offered by Microsoft which allows you to run all of the notebooks contents on a remotely provisioned Azure server).

This setup is recommended for most users -- you won't need to install anything to load these notebooks or to run these commands (though that will take a little more than just opening the files). Thanks,Microsoft!

As I said above, to read these notes, just open the link above and click on the lecture notes. To run one of these notebooks, you will need to create a Microsoft Azure account (we will likely create one of these later anyway, so now is a fine time). once you have created an account and signed in, you should be able to clone any one of these items using the "Clone" button (right above the "Search" bar). this will create a copy in your own personal Microsoft Azure library and will build a server and environment for you any time you launch it.

running locally

if you would like to run these locally, you need to make sure your environment meets the requirements listed in environment.yml. if you are using the conda package manager, you can make sure this is the case by creating an environment based on that environment.yml file.

To create a new environment from scratch, within the directory of this repository run

conda env create -f environment.yml [--name YOUR_NAME]

this will create an environment (named gu511 by default, or YOUR_NAME if you pass your own via the --name command line flag) and install the needed dependencies into that environment.

with some regularity you should also update that created environment to pick up any new dependencies or version updates. if the environment already exists, you can simply run

conda env update -f environment.yml [--name YOUR_NAME]