tau_minus_triplet defined but no triplet STDP about nest-simulator HOT 9 CLOSED

Following up on this:Is the developer module triplet model already available on somebody's fork? Could somebody please make this (even if not fully tested) available?

from nest-simulator.

@abigailm @mhelias

from nest-simulator.

@flinz I have a version of this and I will update it to be mergeable with the current release. A test would be very nice though - do you think you can lend a hand here, to get the model available as soon as possible? There is a SLI test for the basic STDP synapse (tests/unittests/test_stdp_synapse.sli) that could serve as a basis, or you could write a PyNEST test along the same principles - expose the synapse to some pre- and post-synaptic activity, measure the weight, calculate the what the weight should be, compare. Or something else reasonable if you have a better idea.

from nest-simulator.

@abigailm sure, we can take care of testing..

To mention this, a student (@zifeo) is currently reimplementing the Pfister (2006) triplet rule himself, based off the stdp_synapse, which is the reason why i wanted to have a look at the existing implementation, to prevent redundancy.

We'll end up writing tests anyways, which will be equally applicable.

from nest-simulator.

@flinz OK, then perhaps it would be easiest if I get the synapse running with the current release and then send you the source, and you and/or your student make the corresponding PR when you are happy with the testing?

from nest-simulator.

@abigailm sure, this will work!

from nest-simulator.

@flinz I'm on it.

from nest-simulator.

We've opened a PR including the triplet model. #150

from nest-simulator.

Closing, as #150 was merged.

from nest-simulator.