YODATODO Description for the study
TODO stubs for CI banners
- ci/ - scripts and configurations for CI platforms (Travis, Circle-CI, etc)
- code/ - analysis scripts
- docs/ - documentation, notes, papers (pre-prints), etc
- envs/ - images and/or specification of complete environments (singularity, docker, etc)
- inputs/ - dataset(s) which served as inputs for the study
- outputs/ - results of analysis using
code/
,envs/
, andinputs/
- tests/ - various tests to verify correct operation of code etc
- CHANGELOG.md - summary of changes/progress
- HOWTO.md - various details on using/manipulating provided materials
- README.md - high level overview of the study
YODATODO adjust aforementioned list and description below for your particular case/study. Any of the aforementioned directories could be, or could contain, other git/git-annex repositories as submodules. In DataLad they are called subdatasets. Organizing them in such modular fashion allows for reuse while maintaining unambigous version association. Examples:
- a
inputs/bids-dataset
could be used for multiple studies envs/singularity
could point to a collection of singularity images with environments used across multiple studiesoutputs/bids-dataset-preprocessed
could be usedinputs/...
in some other study
- Shablona -
a template project for small scientific Python projects. Could be a good
starting point for your reusable
code/
or a part within it.