metarepo
is short for meta-repository, a GitHub repository that contains instructions to reproduce results in a published work. This repo is a template for creating your own metarepo.
A meta-repository creates a single point of access for someone to find all of the components that were used to create a published work for the purpose of reproducibility. This repository should contain references to all minted data and software as well as any ancillary code used to transform the source data, create figures for your publication, conduct the experiment, and / or execute the contributing software.
your Paper Title here (once published, include a link to the text)
First Last1*, First Last1, and First Last1, 2
1 Pacific Northwest National Laboratory, Richland, WA, USA.
2 Institute for Energy Analysis, Oak Ridge Associated Universities, Washington, DC, USA
* corresponding author: [email protected]
your abstract here
your journal reference
References for each minted software release for all code involved.
These are generated by Zenodo automatically when conducting a release when Zenodo has been linked to your GitHub repository. The Zenodo references are built by setting the author order in order of contribution to the code using the author's GitHub username. This citation can, and likely should, be edited without altering the DOI.
If you have modified a codebase that is outside of a formal release, and the modifications are not planned on being merged back into a version, fork the parent repository and add a .<shortname>
to the version number of the parent and construct your own name. For example, v1.2.5.hydro
.
your software reference here
Reference for each minted data source for your input data. For example:
Human, I.M. (2021). My input dataset name [Data set]. DataHub. https://doi.org/some-doi-number
your input data references here
Reference for each minted data source for your output data. For example:
Human, I.M. (2021). My output dataset name [Data set]. DataHub. https://doi.org/some-doi-number
your output data references here
Model | Version | Repository Link | DOI |
---|---|---|---|
model 1 | version | link to code repository | link to DOI dataset |
model 2 | version | link to code repository | link to DOI dataset |
component 1 | version | link to code repository | link to DOI dataset |
Fill in detailed info here or link to other documentation to thoroughly walkthrough how to use the contents of this repository to reproduce your experiment. Below is an example.
- Install the software components required to conduct the experiment from contributing modeling software
- Download and install the supporting input data required to conduct the experiment
- Run the following scripts in the
workflow
directory to re-create this experiment:
Script Name | Description | How to Run |
---|---|---|
step_one.py |
Script to run the first part of my experiment | python3 step_one.py -f /path/to/inputdata/file_one.csv |
step_two.py |
Script to run the second part of my experiment | python3 step_two.py -o /path/to/my/outputdir |
- Download and unzip the output data from my experiment
- Run the following scripts in the
workflow
directory to compare my outputs to those from the publication
Script Name | Description | How to Run |
---|---|---|
compare.py |
Script to compare my outputs to the original | python3 compare.py --orig /path/to/original/data.csv --new /path/to/new/data.csv |
Use the scripts found in the figures
directory to reproduce the figures used in this publication.
Figure Number(s) | Script Name | Description | How to Run |
---|---|---|---|
1, 2 | generate_plot.py |
Description of figure, ie. "Plots the difference between our two scenarios" | python3 generate_plot.py -input /path/to/inputs -output /path/to/outuptdir |
3 | generate_figure.py |
Description of figure, ie. "Shows how the mean and peak differences are calculated" | python3 generate_figure.py -input /path/to/inputs -output /path/to/outuptdir |