We will be using Jupyter notebooks with the interactive IPython environment to do the labs in this course. This is a good resource on what a jupyter notebook is. Please follow the instructions on this page to set up the software needed to run these jupyter notebooks.
Each week you will be working through 1-3 notebooks during the lab time, and then spending some additional time writing up a short report on what you did.
Getting set up to run these notebooks on your computer can be challenging, so hopefully everything works well for you, but please don’t worry if you have problems getting set up. We have a lot of options and will make sure that everyone can work effectively.
To run these notebooks you are going to install some software on your computer. Depending on the type of computer you have, the details are going to differ a bit, but there are many very good online installation guides.
Please follow these steps:
anaconda (recommended for ease of installation)
miniconda
Backup option: Run on the cloud
Python is a commonly used programming language for doing data analysis in most of physics and astronomy. It is very easy to get started with, so if you are completely new to Python, don’t worry. We won’t be doing anything terribly fancy with python and we will take care to explain the things that we do.
One of the advantages of python is that there is a huge python-based software ecosystem that includes some great data analysis tools. In particular we will be using a few of these python packages in this class.
Depending on how you install python, you might have to install these packages separately. Fortunately, there are some great tools to manage python packages. We will be using the “conda” tool.
The Anaconda data analysis environment is much more complete that miniconda, and includes some other things that you might find useful, but you don’t need the added features for this course. We would recommend installing anaconda if you think you might use python and python packages outside of this course and if you have a lot of disk space.
It does have the nice feature of complete environment, so you won’t have to use the terminal to install anything.
You can download anaconda for free here.
This installation will include numpy, matplotlib, scipy and jupyter so you shouldn’t have to do anything else.
You can also get up and running with Miniconda. Follow the instructions in the link for your computer to set it up. This will require that you do a number of things in the terminal window, but the instructions should be clear enough that you can set things up even if you haven’t used a terminal window before.
Once you have installed and setup up miniconda, run this command in the terminal window to install the extra packages that you will need for this course
`conda install python=3.8 numpy scipy matplotlib jupyter`
The github desktop app is both a nice tool to interface with git, and a way to use git without needing to use a terminal. Here is what you will need to do to use the github desktop app.
git is a software versioning tool. It will allow you to fetch the latest version of the notebook repository to your computer.
You can find instructions on how to install git here.
Once you have installed git, from a terminal window you can do:
git clone https://github.com/KIPAC/Physics89L
to install the repository on your computer.
At any point you can update the repository on your computer to the latest version by doing two commands
git stash
This will move all of the changes that you have made to a backup area so they don’t get overwritten when you get new files for the repository.
git pull
This will fetch all the new changes form the repository and merge them into your version.
If you don’t want to use git, you can always just download a snapshot of the repository each week before class.
Go to https://github.com/KIPAC/Physics89L and click on the green “Code” button in and select “Download ZIP” from the drop-down menu.
Note that this will download the current version of the repository, so you will want to do this each week before class. You probably want to save the older version first so that you can refer to previous work.
If you installed anaconda, then you can start jupyter from the anaconda App graphical interface. (Applications -> Anaconda Navigator, then click on the “Launch” button in the “Jupyter Notebook” box.
One you have started jupyter, it will pop up a browser window showing the contents of what jupyter treats as your home folder. (Either the folder you ran the jupyter-notebook command from, or a folder that you can set in the anaconda preferences). From there you should navigate to the folder that you have installed the course in and click on the notebook you want to run.
If you installed jupyter using miniconda you can start jupyter from a terminal by typing:
jupyter-notebook
To test your setup, navigate to “nb” folder of this repository and run the “Test_Notebook.ipynb” notebook.
If you click on the “Run” button twice, and you have everything set up correctly you should see a plot that looks like this:
You can also run each individual “cell” by clicking on it and typing Shift+Enter.
You can run these notebooks using jupyter on the cloud if you run into issues installing the software on your computer. This uses the service “binder” to run the notebooks on a remote server on the web. This should only be used as a temporary backup option- it has the disadvantage that it can be difficult to save the work you do in the notebook.
You can run the notebooks for week 1 by navigating to this site:
It is fine to run binder the first week, but we would like to get everyone set up with the software on their computers by the second week. This will make it easier for you to save you work and include it in the lab reports. Please contact one of the instructors for help.
