search

Inria-Empenn / narps_open_pipelines

A codebase reproducing the 70 pipelines of the NARPS study (Botvinik-Nezer et al., 2020) shared as an open resource for the community.
MIT License
9 stars 22 forks source link

[ENH] Enhance documentation #171

Open Tritbool opened 7 months ago

Tritbool commented 7 months ago

Goal

Enhance documentation

Why this is useful

VOCABULARY

To help non-expert reproduce the experiment, the vocabulary should be more explicit. in particular we should describe better runs, gain and loss (as values for the subject), and other data that can be misleading because polysemous.

DATA

  1. Data cannot be retrieved using the command cd data; datalad get ./* it misses the --recursive option.
  2. It should be mentioned that the dataset is near 1TB big.
  3. It should be noted that the subjects' data used for the pipelines is determined by a list. Thus, getting only the few first of that list could be sufficient for pipeline implementation and some non-exhaustive testing.

Timeline

N/A

More specifically

No response

Open questions

REPRODUCIBILITY

Maybe we should consider warning users on the RAM needs for the 'worst case' pipeline ?