SU(2) gauge theory with one and two adjoint fermions towards the continuum limit

Analysis workflow

This repository contains the analysis code used to prepare the plots, tables, and other numbers used in the publication SU(2) gauge theory with one and two adjoint fermions towards the continuum limit.

It may be used to regenerate all of the above, starting from raw data outputs from running on HPC, plus data quoted from prior work.

Setup

Getting the code

If you are working from the Zenodo release of this workflow, it is sufficient to download and extract the single archive.

If you are working from the GitHub repository, then to download the workflow and its dependencies, run

shellsession $ git clone --recurse-submodules https://github.com/edbennett/su2_analysis

Software environment

This workflow has been tested on macOS and Linux. The authors are not aware of reasons it should not work on Windows, but this is not tested.

Most dependencies for this workflow are managed via Conda. If you do not already have Conda set up, Miniconda provides the quickest way to get set up.

Additionally, LaTeX is required to be able to produce plots. TeX Live is the best source for this.

With these prerequisites installed, to create a Conda environment to run the workflow, use

shellsession $ conda env create -f environment.yml

(If the default name conflicts with one on your machine, add -n a_name_of_your_choice to the command to override it.)

To activate this environment, run

shellsession $ conda activate su2

(where su2 is replaced with any customised name you have set).

Getting the data

The workflow depends on a number of sources of data:

1. Ensemble metadata. Download the file ensembles.yaml from the data release and place it in the metadata directory.
2. Imported data.
   1. Download the file Fig4.csv from the data release for the publication "Large mass hierarchies from strongly coupled dynamics", and place it in the external_data directory.
   2. Lattice data for $\mathrm{SU}(2)$, $N{\mathrm{f}}=2$, $\beta=2.25$ are included in this repository in the files `externaldata/su2nf2b2.25.csv andexternaldata/su2topology.csv`. No action is needed.
   3. Data output from the previous publication in this project. Download the file su2.sqlite from the data release of the previous work and place it in the external_data directory.
3. Raw data.
   1. Raw data from the previous publication in this project. Download the file raw_data.zip from the data release of the previous work and extract it into the root of the repository.
   2. New raw data prepared for this work. Download the file raw_data.zip from the data release and extract it into the root of the repository.

Running the workflow

With the data described above downloaded, and the environment set up, and the Conda environment active, to run the full analysis end-to-end, run

``` shellsession

python -m autosu2 --ensembles metadata/ensembles.yaml --sideloadcsv externaldata/su2nf2b2.25.csv --sideloadsql externaldata/su2.sqlite

```

This will output all plots, tables, and LaTeX definitions included in the paper into the assets directory. It will also output a CSV of final results (also included in the data release) into the data directory.

Comments

This workflow is the product of seven years of gradual work, and has been built along with the data that it analyses and the research that it enables. As such, it is not necessarily designed to easily generalise to other data, and its overall structure has many shortcomings that would be designed differently were it written from scratch today.

However, should any aspect of the workflow prove useful in your work, please do borrow it, subject to the terms in LICENSE, and making sure to cite this work as specified in CITATION.cff.