/*
run_benchmark.sli
Copyright (C) 2004 The NEST Initiative
This file is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This file is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330,
Boston, MA 02111-1307, USA.
*/
/*
Simulator review
The SLI code in this file creates a sparsely coupled network of
excitatory and inhibitory neurons which exhibits self-sustained
activity after an initial stimulus. Connections within and across
both populations are created at random. Both neuron populations
receive Poissonian background input. The spike output of 500
neurons from each population are recorded. The model is based on
T.P. Vogels & L.F. abbott
Signal Propagation and Logic Gating in Networks of
Integrate-and-Fire Neurons
Journal of Neuroscience, 2005, vol 25, pp 10786-10795.
where neurons are described by the leaky integrate-and-fire
model with conductance based synapses. The synaptic conductances
are decaying exponentials. The Vogels & Abbott model is integrated
on a fixed time grid of 0.1 ms with spike times constrained to this
grid.
The different benchmarks of the FACETS simulator review are variations
of the Vogels & Abbott network model with differences in the model of
the synapse and in the model of the somatic dynamics. The table below
provides an overview
-----------------------------------------------------------------------------
| | synapse model |
# | name | soma | quantity | time course | spike times
-----------------------------------------------------------------------------
1 | coba.sli | i&f | conductance | exponential | grid-constrained
2 | cuba.sli | i&f | current | exponential | grid-constrained
3 | hh_coba.sli | HH | conductance | exponential | grid-constrained
4 | cuba_ps.sli | i&f | current | delta | off-grid
Usage:
This file (run_benchmark.sli) must be used in conjunction with
one of the paranmeter files above, e.g
nest coba.sli run_benchmark.sli
If the number of threads is set to greater than one, the benchmarks can
be run in a distributed fashion using mpirun, e.g.
mpirun -np 2 nest coba.sli runbenchmark.sli
In addition to the four official benchmarks of the FACETS review, a fifth
benchmark has been defined to demonstrate the capabilities of NEST in
simulating large networks with heterogeneous synaptic dynamics. This
benchmark is self contained and can be run from the nest command line
without this file.
-----------------------------------------------------------------------------
5 | cuba_stdp.sli | i&f | stdp-current | exponential | grid-constrained
-----------------------------------------------------------------------------
Marc-Oliver Gewaltig Abigail Morrison, Tobias Potjans
*/
%%% FUNCTION SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Take spike detector, find total number of spikes registered,
% return average rate per neuron in Hz.
% NOTE: If you are running with several MPI processes, this
% function only gives an approximation to the true rate.
%
% spike_det ComputeRate -> rate
/ComputeRate
{
<< >> begin % anonymous dictionary for local variables
/sdet Set % spike detector address
% sum spike numbers across local threads
[0 0 GetStatus /local_num_threads get 1 sub]
{
sdet GetAddress exch append
GetStatus /events get
} Table Plus
/nspikes Set
% We need to guess how many neurons we record from.
% This assumes an even distribution of nodes across
% processes, as well as homogeneous activity in the
% network. So this is really a hack. NEST needs better
% support for rate calculations, such as giving the
% number of neurons recorded from by each spike detector.
Nrec cvd NumProcesses div
/nnrn Set
nspikes nnrn simtime mul div
1000 mul % convert from mHz to Hz, leave on stack
end
} bind % optional, improves performance
def
%%% CONSTRUCTION SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
/BuildNetwork
{
% set global kernel parameters
[0]
<<
/resolution dt
/total_num_virtual_procs virtual_processes
>> SetStatus
tic % start timer on construction
% Set initial parameters for all new neurons and devices
model model_params SetModelStatus
(Creating excitatory population.) = % show message
/E_net model [ NE ] LayoutNetwork def
(Creating inhibitory population.) = % show message
/I_net model [ NI ] LayoutNetwork def
(Creating excitatory stimulus generator.) =
/E_stimulus stimulus Create def
E_stimulus stimulus_params SetStatus
% one detector would in principle be enough,
% but by recording the populations separately,
% we save ourselves a lot of sorting work later
(Creating excitatory spike detector.) =
/E_detector detector Create def
E_detector detector_params SetStatus
(Creating inhibitory spike detector.) =
/I_detector detector Create def
I_detector detector_params SetStatus
% some connecting functions need lists (arrays) over all
% neurons they should work on, so we need to extract these
% lists from the subnetworks
% obtain array with GIDs of all excitatory neurons
/E_neurons E_net GetNodes def
% obtain array with GIDs of all inhibitory neurons
/I_neurons I_net GetNodes def
% all neurons
/allNeurons E_neurons I_neurons join def
/N allNeurons length def
/CE NE epsilon mul iround def %number of incoming excitatory connections
/CI NI epsilon mul iround def %number of incomining inhibitory connections
% number of synapses---just so we know
/Nsyn
CE CI add % internal synapses
N mul
Nrec 2 mul % "synapses" to spike detectors
add
Nstim add % "synapses" from poisson generator
def
% Create custom synapse types with appropriate values for
% our excitatory and inhibitory connections
<< /delay delay >> SetSynapseDefaults
(syn_std) CopySynapsetype ;
E_synapse_params SetSynapseDefaults
(syn_ex) CopySynapsetype ;
I_synapse_params SetSynapseDefaults
(syn_in) CopySynapsetype ;
(Connecting excitatory population.) =
E_neurons
{
/target Set
% use predefined synapse type
syn_ex SetSynapseContext
% E -> E connections
% the following is a single call to RandomConvergentConnect
E_neurons % source population [we pick from this]
target % target neuron
CE % number of source neurons to pick
RandomConvergentConnect
% change synapse type
syn_in SetSynapseContext
% I -> E connections
% as above, but on a single line
I_neurons target CI RandomConvergentConnect
} bind forall % bind improves efficiency
(Connecting inhibitory population.) =
% ... as above, just written more compactly
I_neurons
{
/target Set
syn_ex SetSynapseContext
E_neurons target CE RandomConvergentConnect
syn_in SetSynapseContext
I_neurons target CI RandomConvergentConnect
} bind forall
%Add external stimulus
(Connecting Poisson stimulus.) =
syn_ex SetSynapseContext
E_stimulus
E_neurons Nstim Take % pick the first Nstim neurons
DivergentConnect
% reset synapse type to default
syn_std SetSynapseContext
% Spike detectors are connected to the first Nrec neurons in each
% population. Since neurons are equal and connectivity is homogeneously
% randomized, this is equivalent to picking Nrec neurons at random
% from each population
(Connecting spike detectors.) =
E_neurons Nrec Take % pick the first Nrec neurons
E_detector ConvergentConnect
I_neurons Nrec Take % pick the first Nrec neurons
I_detector ConvergentConnect
% read out time used for building
toc /BuildCPUTime Set
} def
/RunSimulation
{
BuildNetwork
% run, measure computer time with tic-toc
tic
simtime Simulate
toc /SimCPUTime Set
% write a little report
(Simulation summary) =
(Number of Neurons : ) =only N =
(Number of Synapses: ) =only Nsyn =
(Excitatory rate : ) =only E_detector ComputeRate =only ( Hz) =
(Inhibitory rate : ) =only I_detector ComputeRate =only ( Hz) =
(Building time : ) =only BuildCPUTime =only ( s) =
(Simulation time : ) =only SimCPUTime =only ( s\n) =
} def
RunSimulation