Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop #146

flinz · 2015-11-03T15:10:27Z

Addresses issue #82.
Added test for parrot neuron that checks for correct spike repetition with multiplicity.

@lekshmideepu

commit b840820 Merge: 4668ea2 ce63c55 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Tue Nov 3 15:53:36 2015 +0100 Merge branch 'master' into issue-parrot_neuron-multiplicity * master: commit 4668ea2 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Tue Nov 3 15:29:13 2015 +0100 Updated tests to conform to changes in parrot-multipart branch; commit 1b8ddac Merge: bce4bf9 9c4c2f3 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Tue Nov 3 15:27:01 2015 +0100 Merge branch 'parrot-multiport' into issue-parrot_neuron-multiplicity * parrot-multiport: removed unnecessary variables gid_source and gid_parrot; commit bce4bf9 Merge: ef31ab0 3d8da20 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Tue Nov 3 13:20:45 2015 +0100 Merge branch 'parrot-multiport' into issue-parrot_neuron-multiplicity * parrot-multiport: Added tests for STDP protocols between parrot neurons to potentiate and depress connection; Accounted for delays in simulation times; Addresses https://github.com/nest/nest-simulator/pull/118/files#r43736299 and https://github.com/nest/nest-simulator/pull/118/files#r43702948; Fixes https://github.com/nest/nest-simulator/pull/118/files#r43702538 and nest#118 (comment); Import GetStatus/SetStatus from the info component, where they were moved in a3ec4d4. Fix the stop condition of process_static_analysis. Added explanation for recursive import code. Moved Get/SetStatus from node to info component of HL API. Added information on steps to execute after creating the release tarball. Validate parameter values in setters of tsodyks synapses Added empty __init__.py to make the lib directory a module. Split Python high-level API into multiple files and made loading of the submodules dynamic Tsodyks synapse model with common properties commit ef31ab0 Merge: ca70656 9425619 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Tue Nov 3 10:19:25 2015 +0100 Merge branch 'parrot-multiport' into issue-parrot_neuron-multiplicity * parrot-multiport: commit ca70656 Merge: 48a9feb c80a8be Author: Alex Seeholzer <seeholzer@gmail.com> Date: Tue Nov 3 10:18:11 2015 +0100 Merge branch 'master' into issue-parrot_neuron-multiplicity * master: Apply @lekshmideepu suggestions Filter out empty entries in changed files some cleanup Modify travis.yml Implemented two fixes for the K computer: Print status about uploading to S3 Remove cppcheck files after static analysis. Adjust comment and have starting and ending line at summary Correct copyright header. Add a script to parse travis log and give a nice summary. Removed superfluous and outdated files commit 48a9feb Merge: 0b96cd1 4d939e5 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Mon Oct 19 10:14:01 2015 +0200 Merge branch 'parrot-multiport' into issue-parrot_neuron-multiplicity * parrot-multiport: Added superclass (archiving_node) get_status to parrot_neuron; commit 0b96cd1 Author: Alex Seeholzer <seeholzer@gmail.com> Date: Fri Oct 16 16:08:11 2015 +0200 added test for parrot neuron multiplicity. commit ff0cb6d Author: Alex Seeholzer <seeholzer@gmail.com> Date: Fri Oct 16 15:39:45 2015 +0200 Changed parrot neuron to emit multiple spikes via spike multiplicity instead of a loop; Addresses nest#82

heplesser · 2015-11-03T22:53:54Z

Looks good, I created a PR to @flinz repo with improved documentation and one more test.

heplesser · 2015-11-03T23:30:41Z

models/parrot_neuron.cpp

+      SpikeEvent se;
+      se.set_multiplicity( current_spikes_n );
+
+      network()->send( *this, se, lag );
      set_spiketime( Time::step( origin.get_steps() + lag + 1 ) );


I need to double-check whether set_spiketime() needs to be called once per outgoing spike, i.e., as many times as multiplicity indicates.

#77 (comment) seems to suggest that this is indeed the case.

this would result in

for (ulong_t i=0; i<current_spikes_n; i++ ) set_spiketime( Time::step( origin.get_steps() + lag + 1 ) );

with which we're almost back to the old form -- although it's a little cleaner with the handling of multiplicity.

It should be noted that the original model did not set the spike_times multiply -- but the parrot probably wasn't used for plasticity much yet.

I agree that you need this. Can you please add this change?

@flinz I agree with @jougs and after more meditation over the spike archiving and plasticity (see #77) think that we do have to call set_spiketime() once for every spike. One this PR is merged, I will extend set_spiketime() to accept multiplicity as argument in a separate PR, but I don't want to delay this PR further.

…-squashed Revised documentation of parrot_neuron and added one more test.

jougs · 2015-11-04T12:49:03Z

models/parrot_neuron.h

 No parameters to be set in the status dictionary.

-References:
-No references
-
 Author:  May 2006, Reichert, Morrison


This should now be

Author: David Reichert, Abigail Morrison, Alexander Seeholzer, Hans Ekkehard Plesser FirstVersion: May 2006

jougs · 2015-11-04T12:50:51Z

Very nice work. I'm all 👍 once my comments above have been addressed.

heplesser · 2015-11-04T15:31:09Z

👍 once the loop around set_multiplicity() is in place and Jochen's comments are addressed.

* master: Use NewConnect instead of DivergentConnect and remove non-default weight in favor of higher rate. Fixed error message and reporting in case CSA is not available. Make wording more consistent with respect to active and passive speech. improvements as discussed in PR nest#127 Added reference Updated CSA examples to new documentation guidelines. Replace NAN, std::isnan by numerics::nan, numerics::isnan

…quashed Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop

# Conflicts: # models/parrot_neuron.cpp

flinz changed the title ~~Changed parrot neuron to emit multiple spikes via spike multiplicity instead of a loop~~ Fix issue https://github.com/nest/nest-simulator/issues/82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop Nov 3, 2015

flinz changed the title ~~Fix issue https://github.com/nest/nest-simulator/issues/82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop~~ Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop Nov 3, 2015

Revised documentation of parrot_neuron and added one more test.

87eb6b7

heplesser reviewed Nov 3, 2015
View reviewed changes

Merge pull request #1 from heplesser/issue-parrot_neuron-multiplicity…

7f8f8f5

…-squashed Revised documentation of parrot_neuron and added one more test.

jougs reviewed Nov 4, 2015
View reviewed changes

flinz added 2 commits November 4, 2015 17:18

Spike_times are now set with multiplicity. Author changes in docstring.

fc48425

jougs added a commit that referenced this pull request Nov 5, 2015

Merge pull request #146 from flinz/issue-parrot_neuron-multiplicity-s…

ca0513f

…quashed Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop

jougs merged commit ca0513f into nest:master Nov 5, 2015

flinz deleted the issue-parrot_neuron-multiplicity-squashed branch November 6, 2015 11:57

tammoippen added a commit to tammoippen/nest-simulator that referenced this pull request Jan 7, 2016

Merge commit 'ca0513f8' into nest3 (PR nest#146)

f0c51eb

# Conflicts: # models/parrot_neuron.cpp

gewaltig mentioned this pull request Jan 12, 2016

Refactoring NEST #205

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop #146

Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop #146

flinz commented Nov 3, 2015

heplesser commented Nov 3, 2015

heplesser Nov 3, 2015

flinz Nov 4, 2015

jougs Nov 4, 2015

heplesser Nov 4, 2015

jougs Nov 4, 2015

jougs commented Nov 4, 2015

heplesser commented Nov 4, 2015

Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop #146

Fix issue #82: parrot neuron now emits multiple spikes via spike multiplicity instead of a loop #146

Conversation

flinz commented Nov 3, 2015

heplesser commented Nov 3, 2015

heplesser Nov 3, 2015

Choose a reason for hiding this comment

flinz Nov 4, 2015

Choose a reason for hiding this comment

jougs Nov 4, 2015

Choose a reason for hiding this comment

heplesser Nov 4, 2015

Choose a reason for hiding this comment

jougs Nov 4, 2015

Choose a reason for hiding this comment

jougs commented Nov 4, 2015

heplesser commented Nov 4, 2015