MRG: thinking of an interface for HRF convolution #290

merged 2 commits into from Oct 1, 2015


None yet
1 participant

matthew-brett commented Sep 1, 2013

I'm trying to work out a good interface to express the fact that generally,
when we convolve things with HRFs, we want the same interval for the kernel,
and the same dt.

Maybe we could have an HRF object, that has a 'convolve' method, maybe an
'evaluate' method (to evaluation the real function at given time values, and a
symfunc property for the symbolic function.

I haven't really thought this last part through.


matthew-brett commented Sep 25, 2015

Any comments here? I think this code is an improvement, and we can merge this and think of the final interface for the convolver class later.

matthew-brett changed the title from WIP: thinking of an interface for HRF convolution to MRG: thinking of an interface for HRF convolution Sep 25, 2015


matthew-brett commented Sep 29, 2015

Last call for comments.

matthew-brett added some commits Aug 31, 2013

@matthew-brett matthew-brett NF: add and test TimeConvolver class
TimeConvolver wraps a function of time, a support interval and a time
delta to make it simpler and more efficient to convolve time courses.
@matthew-brett matthew-brett TST: add use of TimeConvolver for HRFs
Test TimeConvolver again aganst convolve_functions, and summing of
individual convolved events.

@matthew-brett matthew-brett added a commit that referenced this pull request Oct 1, 2015

@matthew-brett matthew-brett Merge pull request #290 from matthew-brett/convolve-kernels
MRG: thinking of an interface for HRF convolution

Add convolver class, as a step for thinking about a higher-level HRF object.

Closes #16

@matthew-brett matthew-brett merged commit 0fc0e30 into nipy:master Oct 1, 2015

2 checks passed

continuous-integration/travis-ci/pr The Travis CI build passed
coverage/coveralls Coverage increased (+0.02%) to 86.771%
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment