CHIME analysis tutorial

This repo contains a collection of notebooks and scripts intended to go over some of the basics of CHIME data analysis and pipeline development.

Setting up a CHIME environment

Whether you are running this on cedar, or on your own laptop you'll need to setup a Python environment with the CHIME pipeline code.

First you'll need a base Python environment. On cedar this can be done by loading the CHIME python modules using

module use /project/rpp-chime/chime/chime_env/modules/modulefiles/
module load chime/python/2022.06

either on the command line or in your .bashrc (or .zshrc). On another machine you're on your own, but you can either do this via your system package manager (apt, homebrew etc.), or via a distribution like Anaconda or Miniforge.

With a working Python installation you'll need to have an install of the CHIME pipeline packages. On cedar you can get a quick installation going by loading the chime/pipeline module using

module load chime/pipeline/2022.09

However, if you're going to be developing you'll want your own personal copy. We strongly recommend that you use a virtualenv to do this, so much so that we have written a script to automate this for you. Simply run

$ mkchimeenv create <install dir>

and it will create a virtual environment in <install dir>/venv/ and install all CHIME packages into it as editable git repositories (available at <install dir>/code/). To activate this environment simply activate it as usual source <install dir>/venv/bin/activate. On cedar the mkchimeenv command is already installed, on another machine you can install it from this repository.

The final and optional step is to install a kernel such that you can run this virtual environment in Jupyter. To do this activate your virtual environment as described above and then run python -m ipykernel install --user --name <env name>, where <env name> is a name of your choice that you will use to refer to it. You may need to write a kernel wrapper.

Sample Data

Some of the tutorials require sample data. This is all available from the CHIME website.

Tutorials

• ringmaps/: how to load and interpret a ringmap.
• chime_ringmaps/: how to load and manipulate a real ringmap on cedar
• dataset_ids/: how to understand and interpret the dataset_ids and fetch the relevant states from the database.
• baseband/: decode kotekan baseband dumps for RFI testing.