the pipeline model is used to fit and predict X and y variables that weren't declared before in the tutorial:

The idea of this section being that more focussed on how to do feature selection procedure providing an idea. For me, I think user will know the meaning of X and y in fit and predict (as it explained above).

In the first paragraph of section 2.4.1, both links to [Grosenick et al. 2013] redirect to [Gramfort et al. 2013], they should probably be changed to http://www.ncbi.nlm.nih.gov/pubmed/23298747.

Thanks for spotting. PR welcome. :)

Ok, I'll do a PR

On Tue, Jun 14, 2016 at 11:05 PM, Joao Loula notifications@github.com
wrote:

In section 2.1.3.1,
http://nilearn.github.io/decoding/decoding_tutorial.html#dimension-reduction-with-feature-selection
the pipeline model is used to fit and predict X and y variables that
weren't declared before in the tutorial: it seems these were taken from the
example plot_haxby_anova_svm.py, so something like this should probably
be added:

mask_filename = haxby_dataset.mask
nifti_masker = NiftiMasker(mask_img=mask_filename, standardize=True)
func_filename = haxby_dataset.func[0]
X = nifti_masker.fit_transform(func_filename)
X = X[condition_mask]

with eventually a change in the name of the mask variables just to make it
clear that they're different from those used at the beginning of the
section. The y variable can just be replaced with the target variable
declared before.

In the first paragraph of section 2.4.1
http://nilearn.github.io/decoding/space_net.html#the-spacenet-decoder,
both links to [Grosenick et al. 2013] redirect to [Gramfort et al. 2013],
they should probably be changed to
http://www.ncbi.nlm.nih.gov/pubmed/23298747.

Good catch.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#1130, or mute the thread
https://github.com/notifications/unsubscribe/AAms1Ia4nL3lNUBaf2xnNnP68LD2PxESks5qLxevgaJpZM4I1wqY
.

Elvis Dohmatob,
sent from my potato...
http://dohmatob.blogspot.fr/, https://team.inria.fr/parietal/elvis/,
https://github.com/dohmatob

It looks like the link issues were resolved in #1184, so now the only problem is that Example 2.1 (An introduction to decoding) does not define the variables used in section 2.1.3.1 (Dimension reduction with feature selection)- specifically func_img and conditions.

The example uses the following lines from plot_haxby_anova_svm.py, which does not include the code used to define func_img and conditions: