Felix G. Knorr

I see. Are there instructions somewhere on how to add them? If it's doable for me, and not requiring tons of winapi expertise, I'd try to add it myself.

I see. Thanks for the info. It seems that header isn't even part of mingw64, and its neccessary to use vcc. But since its a windows project anyway that's no...

I will try at least ^^

Sorry for reopening, but to me, this seemed more appropriate than opening a new issue. I tried to write a wrapper for the API, but I've hit a wall. When...

You were right, making this change will make the api seem to work (after also changing GetActiveWindow() to GetForegroundWindow()). However, it outputs "false" in the end. And as far as...

Currently, I work with the fmri of subjects that was recorded during a task switching paradigm, and I use the SpaceNet classifier to classify which task they execute. The Idea...

I don't really use the SpaceNet on time series data, I apply a GLM (SPM12) block wise and use the resulting beta-coefficient maps as input for the SpaceNet. I then...

I think I can contribute the fix. The question is what to do exactly. I could simply change the behavior, but since u mentioned changing parameters, I'm not clear on...

But then this would remove the psc option. Also, since you don't seem too concerned with keeping the current way in there, I'd propose to let the parameters as they...

Ok, this code represents the problem: ```python clf = SpaceNetClassifier() clf.fit(train_niis) half = len(test_niis) // 2 distances1 = clf.decision_function(clf.masker_.transform(test_niis))[:half] distances2 = clf.decision_function(clf.makser_.transform(test_niis[:half])) assert np.allclose(distances1, distances2) # should pass, but will...