bthirion

Document details of design matrix generation

1

Describe (at least) `overampling` and `min_onset`

Currently, the occurrence of different sets of trial types across sessions in a FirstLevelGLM raises an error, because the model expects the conditions to be identical in all sessions. This...

Image segmentation needs more testing

4

The example in examples/tissue_classification.py does not work e.g. on the following image: nipy/data/neurospin/sulcal2000/nobias_ammon.nii.gz

Convergence of algorithms.graph with networkx

3

## A common/similar API The algorithms.graph library is quite comparable to the specification of a nx.DiGraph object, with some restrictions, plus some nx algorithms. The function-based structure of nx is...

Simply replace `label_img` with `maps_img` in the docstring ### Affected pages https://github.com/nilearn/nilearn/blob/main/nilearn/plotting/find_cuts.py

DecoderRegressor API is not very clear

1

This is a matter of ergonomy rather than a bug, but when using `DecoderRegressor`, the computed `cv_scores_` attribute corresponding to folds come out as a dictionary with key `beta`, which...

Rank-Deficient second-level design matrices should be allowed

1

It turns out that the parametrization of second-level models often leads to rank-deficient design matrices. We should make this legal. The only issue is that some contrasts are estimable while...

I find it very important that examples are visual.

Example_compare_mutual_info_correlation.py

1

print ___doc___ at the beginning of the example I would show the samples in a scatter plot to make it clear that you are sampling points from a dataset. Less...

I'd rather put the simulated_data in the library, e.g. in a '_utils.py' module, otherwise, people will not understand what this stands for.