Alejandro de la Vega
Alejandro de la Vega
The [ancpbids](https://github.com/ANCPLabOldenburg/ancp-bids) project has made formidable progress in implementing a `BIDSLayout`-like API, with marked performance improvements, using a different underlying implementation with a custom query language. Given limited resources to...
Given that for simple models (i.e. contrasts at run level, and dummy contrasts at subject and dataset level) `GroupBy` must be set, I can't imagine a situation where letting pybids...
In a scenario where subjects have 1-2 runs (with those w/ only 1 run having no run labels), `Analysis.setup()` crashes when the statsmodel doesn't define `Run` under `Input`: ``` ERROR:nipype.workflow:could...
To address loading incompatible saved dbs from previous various we could: - Throw a warning when loading a db from a previous pybids version - Catch SQL exceptions and suggest...
Currently, if I want to have an analysis run for subject 1, run 1 and subject 2 run 2, it's not possible as the following specified subject 1+2 runs 1+2,...
``` In [1]: from bids.layout import BIDSLayout In [2]: layout = BIDSLayout('/datasets/raw/SherlockMerlin/', derivatives=True) In [3]: layout.get_collections(subject='19', task='MerlinMovie', level='run') ``` Throws the following error: ``` /usr/local/lib/python3.6/site-packages/bids/layout/layout.py in get_collections(self, level, types, variables,...
I pass an ignore directing to `BIDSLayout` in order to not index fmriprep confounds: ``` layout = BIDSLayout('/work/HealthyBrainNetwork/preproc/fmriprep', derivatives=['/work/HealthyBrainNetwork/neuroscout-bundles/MzXpw', '/work/HealthyBrainNetwork/preproc/fmriprep'], ignore=[re.compile('.*desc-confounds_regressors.tsv')]) ``` However, when I make my analysis, and ask...
To facilitate with the adoption of BIDS Stats Model, we can developed a CLI wizard to create StatsModels. If this is useful, we could potentially also upstream these changes into...
Since 0.8, the API has changed, such as the deprecation of `get_design_matrix`. This is not reflected in the docs, yet.