[documentation] Activating the coreNeuron package in a NetPyNE example

jhielson commented 3 years ago

Hi all,

I was wondering if there is a tutorial explaining how to use coreNeuron on a NetpyNE model. I have installed the coreNeuron package already. But, based on their documentation, the following commands are necessary to activate the GPU support:

Enable cache efficiency :

from neuron import h h.cvode.cache_efficient(1)
Enable CoreNEURON :

from neuron import coreneuron coreneuron.enable = True
If GPU support is enabled during build, enable GPU execution using:

coreneuron.gpu = True

:warning: In this case you must launch your script using the special binary! This is explained in more detail below.
Use psolve to run simulation after initialization :

h.stdinit() pc.psolve(h.tstop)

Is there anything similar for NetPyNE?

iomaganaris commented 2 years ago

Hello @jhielson ,

The change to enable CoreNEURON through NetPyNE is now merged. You can enable CoreNEURON and GPU execution using the following options in your cfg.py file:

cfg.coreneuron = True
cfg.gpu = True

You can find an example in https://github.com/Neurosim-lab/netpyne/blob/development/examples/M1detailed/cfg.py. Let us know if you have any issue with this

jhielson commented 2 years ago

Dear @iomaganaris,

Thanks for your update. I am not sure if it is working in my computer yet. First, I cloned the development branch of the netpyne repository and installed it in my computer.

$git clone https://github.com/Neurosim-lab/netpyne.git
$cd netpyne
$git checkout development
$pip install -e .

Then, I added those lines to my code as presented in the example file.

   def buildSimConfig(self):
        # Simulation options
        simConfig = specs.SimConfig()            
        simConfig.duration = self.t_sim            
        simConfig.dt = 2                   
        simConfig.coreneuron = True                    #<<<<<<<<<<<<<
        simConfig.gpu = True                                #<<<<<<<<<<<<<
        ...

As a result, my simulation runs fine without any error message. But the run time and real-time ratio has not been affected. I was expecting an improvement on that. What do you think?

iomaganaris commented 2 years ago

Hello @jhielson ,

My guess about this is issue that you didn't use a NEURON installation with CoreNEURON enabled. To be able to run CoreNEURON on GPU transparently with NEURON you have to install NEURON with CoreNEURON from source using these instructions. After installing NEURON using these instructions for a GPU enabled build you will be able to compile your mod files with <install_directory>/bin/nrnivmodl -coreneuron <modfile_directory>. Then this command will generate a x86_64 folder if you're using a system with an x86_64 architecture processor. Otherwise this folder will have the processor architecture name. Inside this folder there will be a special executable that is linked with the CoreNEURON libraries needed to run the simulation on GPU. You may launch NetPyNE now using the following command:

cd <the_folder_where_your_init.py_file_for_netpyne_exists>
mpirun -n <num_procs> ./x86_64/special -mpi -python init.py

After the model building stage of your simulation you should see some extra logs coming from the CoreNEURON simulation that informs you about how many GPUs and threads/ranks you used for the simulation as well as the simulation runtime that contributes to the real-time ratio. Please let us know whether this was the issue and you manage to run your simulation using the above instructions. Thank you.

jhielson commented 2 years ago

Hi Ioannis,

I apologise for the delayed response. My mon was hospitalised this weekend and I have been giving her some support. I will follow your instructions later next week. I appreciate your help.

Best regards, Jhielson Montino Pimentel

Get Outlook for Androidhttps://aka.ms/AAb9ysg

From: Ioannis Magkanaris @.***> Sent: Monday, 6 December 2021, 07:38 To: Neurosim-lab/netpyne Cc: MONTINO PIMENTEL Jhielson; Mention Subject: Re: [Neurosim-lab/netpyne] [documentation] Activating the coreNeuron package in a NetPyNE example (#611)

This email was sent to you by someone outside the University. You should only click on links or attachments if you are certain that the email is genuine and the content is safe.

Hello @jhielsonhttps://github.com/jhielson ,

My guess about this is issue that you didn't use a NEURON installation with CoreNEURON enabled. To be able to run CoreNEURON on GPU transparently with NEURON you have to install NEURON with CoreNEURON from source using these instructionshttps://github.com/BlueBrain/CoreNeuron#installation. After installing NEURON using these instructions for a GPU enabled build you will be able to compile your mod files with /bin/nrnivmodl -coreneuron . Then this command will generate a x86_64 folder if you're using a system with an x86_64 architecture processor. Otherwise this folder will have the processor architecture name. Inside this folder there will be a special executable that is linked with the CoreNEURON libraries needed to run the simulation on GPU. You may launch NetPyNE now using the following command:

cd mpirun -n ./x86_64/special -mpi -python init.py

After the model building stage of your simulation you should see some extra logs coming from the CoreNEURON simulation that informs you about how many GPUs and threads/ranks you used for the simulation as well as the simulation runtime that contributes to the real-time ratio. Please let us know whether this was the issue and you manage to run your simulation using the above instructions. Thank you.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/Neurosim-lab/netpyne/issues/611#issuecomment-986650362, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AAQ66KBS3WOTUOUQDCLXVDDUPSHB5ANCNFSM47OLO7RQ. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. Is e buidheann carthannais a th’ ann an Oilthigh Dhùn Èideann, clàraichte an Alba, àireamh clàraidh SC005336.

iomaganaris commented 2 years ago

Hello Jhielson,

I'm sorry to hear that and I wish a speedy recovery. Take your time and let me know if you need any help whenever you work on this.

Kind regards, Ioannis

jhielson commented 2 years ago

Dear @iomaganaris,

Thank you for your support. It seems that the coreneuron package is working fine now. I can compile the mod files in my project folder and it generates the /x86_64 folder with a special executable file on it. However, after running my simulation via mpirun command, I get the following error:

$mpirun -n 4 ./x86_64/special -mpi -python cerebellumCells.py


numprocs=4
NEURON -- VERSION 8.0a-590-ge40208d8+ master (e40208d8+) 2021-06-18
Duke, Yale, and the BlueBrain Project -- Copyright 1984-2021
See http://neuron.yale.edu/neuron/credits

Additional mechanisms from files
 "spikeout.mod" "stdp.mod" "SynExp2NMDA.mod"

Start time:  2022-01-24 12:53:01.201457

Creating network of 109 cell populations on 4 hosts...
  Number of cells on node 1: 27 
  Number of cells on node 2: 27 
  Number of cells on node 3: 27 
  Number of cells on node 0: 28 
  Done; cell creation time = 0.03 s.
Making connections...
  Number of connections on node 1: 176 
  Number of connections on node 0: 176 
  Number of connections on node 3: 176 
  Number of connections on node 2: 176 
  Done; cell connection time = 0.13 s.
Adding stims...
  Number of stims on node 2: 23 
  Number of stims on node 1: 23 
  Number of stims on node 0: 24 
  Number of stims on node 3: 23 
  Done; cell stims creation time = 0.01 s.
Recording 27 traces of 1 types on node 1
Recording 27 traces of 1 types on node 2
Recording 27 traces of 1 types on node 3
Recording 28 traces of 1 types on node 0
0 ./x86_64/special: usable mindelay is 0 (or less than dt for fixed step method)
0  near line 0
0  tstop=1000.0
             ^
        0 finitialize(-65)
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
with errorcode -1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

Have you seen this error message before?

usable mindelay is 0 (or less than dt for fixed step method)

Best regards, Jhielson M. Pimentel

jhielson commented 2 years ago

Dear @iomaganaris,

After reducing the 'dt' to 0.025ms, the simulation was executed without errors. However, I am not sure if the performance of the simulation was optimized.

You can find the output of my simulation below:

numprocs=4
NEURON -- VERSION 8.0a-590-ge40208d8+ master (e40208d8+) 2021-06-18
Duke, Yale, and the BlueBrain Project -- Copyright 1984-2021
See http://neuron.yale.edu/neuron/credits

Additional mechanisms from files
 "spikeout.mod" "stdp.mod" "SynExp2NMDA.mod"

Start time:  2022-01-26 15:10:55.549057

Creating network of 109 cell populations on 4 hosts...
  Number of cells on node 1: 27 
  Number of cells on node 2: 27 
  Number of cells on node 3: 27 
  Number of cells on node 0: 28 
  Done; cell creation time = 0.03 s.
Making connections...
  Number of connections on node 0: 176 
  Number of connections on node 2: 176 
  Number of connections on node 1: 176 
  Number of connections on node 3: 176 
  Done; cell connection time = 0.14 s.
Adding stims...
  Number of stims on node 3: 23 
  Number of stims on node 0: 24 
  Number of stims on node 2: 23 
  Number of stims on node 1: 23 
  Done; cell stims creation time = 0.01 s.
Recording 27 traces of 1 types on node 2
Recording 27 traces of 1 types on node 1
Recording 27 traces of 1 types on node 3
Recording 28 traces of 1 types on node 0

Running with interval func  ...

Gathering data...
  Done; gather time = 0.04 s.

Analyzing...
  Cells: 109
  Connections: 704 (6.46 per cell)
  Spikes: 0 (0.00 Hz)
  Simulated time: 10.0 s; 4 workers
  Run time: 1643209855.76 s

... ...

Analyzing...
  Cells: 109
  Connections: 704 (6.46 per cell)
  Spikes: 0 (0.00 Hz)
  Simulated time: 1.0 s; 4 workers
  Run time: 1643210117.02 s
  Done; run time = 2.98 s; real-time ratio: 0.34.

The real-time ratio was not good. I can use up to 6 workers but the difference is not much. How can know that the GPU is being really being used?

iomaganaris commented 2 years ago

Dear @jhielson,

I'm glad you fixed your first issue and I'm sorry for my late reply. I think that the issue in your run is that CoreNEURON was not enabled because it seems like you call https://github.com/suny-downstate-medical-center/netpyne/blob/88a565ceceffae75f86c839256e90adaa4584301/netpyne/sim/run.py#L175 from NetPyNE which doesn't have CoreNEURON support at the moment. We'll have to take a look into that and come back to you.

jhielson commented 2 years ago

Thanks, @iomaganaris. Indeed, I have been using the "runSimWithIntervalFunc" function in my code. Actually, that one is quite important since it allows me to interact with my model during simulation.

jhielson commented 2 years ago

Dear @iomaganaris,

I hope you are doing well. I have decided to incorporate an Izhikevich model into my code but it seems this model (izhi2007b.mod) is incompatible with CoreNEURON. Do you know how to solve this issue?

jhielson commented 2 years ago

This is the output of my compilation (issues with izhi2007b.mod and stdp.mod):

Mod files: "./izhi2007b.mod" "./spikeout.mod" "./stdp.mod" "./SynExp2NMDA.mod"

COBJS=''
[INFO] Running: make -j4 -f /usr/share/coreneuron/nrnivmodl_core_makefile ROOT=/usr MODS_PATH=x86_64/corenrn/mod2c BUILD_TYPE=STATIC
Default NMODL flags: 
Translating x86_64/corenrn/mod2c/stdp.mod into x86_64/corenrn/mod2c//stdp.cpp
Translating x86_64/corenrn/mod2c/izhi2007b.mod into x86_64/corenrn/mod2c//izhi2007b.cpp
Notice: Use of POINTER is not thread safe.
Notice: VERBATIM blocks are not thread safe
cvode_emit=0 cvode_not_allowed=0
Error : x86_64/corenrn/mod2c/izhi2007b.mod is not thread safe and incompatible with CoreNEURON
See details at https://neuron.yale.edu/neuron/docs/multithread-parallelizationcvode_emit=0 cvode_not_allowed=0
Error : x86_64/corenrn/mod2c/stdp.mod is not thread safe and incompatible with CoreNEURON
See details at https://neuron.yale.edu/neuron/docs/multithread-parallelizationAborted (core dumped)
/usr/share/coreneuron/nrnivmodl_core_makefile:225: recipe for target 'x86_64/corenrn/mod2c/izhi2007b.cpp' failed
make: *** [x86_64/corenrn/mod2c/izhi2007b.cpp] Error 134
make: *** Deleting file 'x86_64/corenrn/mod2c/izhi2007b.cpp'
make: *** Waiting for unfinished jobs....
Aborted (core dumped)
/usr/share/coreneuron/nrnivmodl_core_makefile:225: recipe for target 'x86_64/corenrn/mod2c/stdp.cpp' failed
make: *** [x86_64/corenrn/mod2c/stdp.cpp] Error 134
make: *** Deleting file 'x86_64/corenrn/mod2c/stdp.cpp'

iomaganaris commented 2 years ago

Hello @jhielson ,

Could you please share which mod files exactly are you using? I guess you're referring to the izhi2007b.mod from https://senselab.med.yale.edu/ModelDB/ShowModel?model=39948? Thank you!

jhielson commented 2 years ago

Hi @iomaganaris,

Exactly, that one. The other one is the 'STDP.mod' but I might not use it in my code.

stdp.mod

COMMENT

STDP + RL weight adjuster mechanism

Original STDP code adapted from:
http://senselab.med.yale.edu/modeldb/showmodel.asp?model=64261&file=\bfstdp\stdwa_songabbott.mod

Adapted to implement a "nearest-neighbor spike-interaction" model (see 
Scholarpedia article on STDP) that just looks at the last-seen pre- and 
post-synaptic spikes, and implementing a reinforcement learning algorithm based
on (Chadderdon et al., 2012):
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0047251

Example Python usage:

from neuron import h

## Create cells
dummy = h.Section() # Create a dummy section to put the point processes in
ncells = 2
cells = []
for c in range(ncells): cells.append(h.IntFire4(0,sec=dummy)) # Create the cells

## Create synapses
threshold = 10 # Set voltage threshold
delay = 1 # Set connection delay
singlesyn = h.NetCon(cells[0],cells[1], threshold, delay, 0.5) # Create a connection between the cells
stdpmech = h.STDP(0,sec=dummy) # Create the STDP mechanism
presyn = h.NetCon(cells[0],stdpmech, threshold, delay, 1) # Feed presynaptic spikes to the STDP mechanism -- must have weight >0
pstsyn = h.NetCon(cells[1],stdpmech, threshold, delay, -1) # Feed postsynaptic spikes to the STDP mechanism -- must have weight <0
h.setpointer(singlesyn._ref_weight[0],'synweight',stdpmech) # Point the STDP mechanism to the connection weight

Version: 2013oct24 by cliffk

ENDCOMMENT

NEURON {
    POINT_PROCESS STDP : Definition of mechanism
    POINTER synweight : Pointer to the weight (in a NetCon object) to be adjusted.
    RANGE tauhebb, tauanti : LTP/LTD decay time constants (in ms) for the Hebbian (pre-before-post-synaptic spikes), and anti-Hebbian (post-before-pre-synaptic) cases. 
    RANGE hebbwt, antiwt : Maximal adjustment (can be positive or negative) for Hebbian and anti-Hebbian cases (i.e., as inter-spike interval approaches zero).  This should be set positive for LTP and negative for LTD.
    RANGE RLwindhebb, RLwindanti : Maximum interval between pre- and post-synaptic events for an starting an eligibility trace.  There are separate ones for the Hebbian and anti-Hebbian events.
    RANGE useRLexp : Use exponentially decaying eligibility traces?  If 0, then the eligibility traces are binary, turning on at the beginning and completely off after time has passed corresponding to RLlen.
    RANGE RLlenhebb, RLlenanti : Length of the eligibility Hebbian and anti-Hebbian eligibility traces, or the decay time constants if the traces are decaying exponentials.
    RANGE RLhebbwt, RLantiwt : Maximum gains to be applied to the reward or punishing signal by Hebbian and anti-Hebbian eligibility traces.  
    RANGE wmax : The maximum weight for the synapse.
    RANGE softthresh : Flag turning on "soft thresholding" for the maximal adjustment parameters.
    RANGE STDPon : Flag for turning STDP adjustment on / off.
    RANGE RLon : Flag for turning RL adjustment on / off.
    RANGE verbose : Flag for turning off prints of weight update events for debugging.
    RANGE tlastpre, tlastpost : Remembered times for last pre- and post-synaptic spikes.
    RANGE tlasthebbelig, tlastantielig : Remembered times for Hebbian anti-Hebbian eligibility traces.
    RANGE interval : Interval between current time t and previous spike.
    RANGE deltaw : The calculated weight change.
    RANGE newweight : New calculated weight.
    RANGE skip : Flag to skip 2nd set of conditions
}

ASSIGNED {
    synweight        
    tlastpre   (ms)    
    tlastpost  (ms)   
    tlasthebbelig   (ms)    
    tlastantielig  (ms)        
    interval    (ms)    
    deltaw
    newweight          
}

INITIAL {
    tlastpre = -1            : no spike yet
    tlastpost = -1           : no spike yet
    tlasthebbelig = -1      : no eligibility yet
    tlastantielig = -1  : no eligibility yet   
    interval = 0
    deltaw = 0
    newweight = 0
}

PARAMETER {
    tauhebb  = 10  (ms)   
    tauanti  = 10  (ms)    
    hebbwt = 1.0
    antiwt = -1.0
    RLwindhebb = 10 (ms)
    RLwindanti = 10 (ms)
    useRLexp = 0   : default to using binary eligibility traces
    RLlenhebb = 100 (ms)
    RLlenanti = 100 (ms)
    RLhebbwt = 1.0
    RLantiwt = -1.0
    wmax  = 15.0
    softthresh = 0
    STDPon = 1
    RLon = 1
    verbose = 0
    skip = 0
}

NET_RECEIVE (w) {
    deltaw = 0.0 : Default the weight change to 0.
    skip = 0

    if (verbose > 0)  { printf("t=%f (BEFORE) tlaspre=%f, tlastpost=%f, flag=%f, w=%f, deltaw=%f \n",t,tlastpre, tlastpost,flag,w,deltaw) }

    : Hebbian weight update happens 1ms later to check for simultaneous spikes (otherwise bug when using mpi)
    if ((flag == -1) && (tlastpre != t-1)) {   
        skip = 1 : skip the 2nd set of conditions since this was artificial net event to update weights
        deltaw = hebbwt * exp(-interval / tauhebb) : Use the Hebbian decay to set the Hebbian weight adjustment. 
        if (softthresh == 1) { deltaw = softthreshold(deltaw) } : If we have soft-thresholding on, apply it.
        adjustweight(deltaw) : Adjust the weight.
        if (verbose > 0) { printf("Hebbian STDP event: t = %f ms; tlastpre = %f; w = %f; deltaw = %f\n",t,tlastpre,w,deltaw) } : Show weight update information if debugging on.
        }

    : Ant-hebbian weight update happens 1ms later to check for simultaneous spikes (otherwise bug when using mpi)
    else if ((flag == 1) && (tlastpost != t-1)) { :update weight 1ms later to check for simultaneous spikes (otherwise bug when using mpi)
        skip = 1 : skip the 2nd set of conditions since this was artificial net event to update weights
        deltaw = antiwt * exp(interval / tauanti) : Use the anti-Hebbian decay to set the anti-Hebbian weight adjustment.
        if (softthresh == 1) { deltaw = softthreshold(deltaw) } : If we have soft-thresholding on, apply it.
        adjustweight(deltaw) : Adjust the weight.
        if (verbose > 0) { printf("anti-Hebbian STDP event: t = %f ms; deltaw = %f\n",t,deltaw) } : Show weight update information if debugging on. 
        }

    : If we receive a non-negative weight value, we are receiving a pre-synaptic spike (and thus need to check for an anti-Hebbian event, since the post-synaptic weight must be earlier).
    if (skip == 0) {
        if (w >= 0) {           
            interval = tlastpost - t  : Get the interval; interval is negative
            if  ((tlastpost > -1) && (-interval > 1.0)) { : If we had a post-synaptic spike and a non-zero interval...
                if (STDPon == 1) { : If STDP learning is turned on...
                    if (verbose > 0) {printf("net_send(1,1)\n")}
                    net_send(1,1) : instead of updating weight directly, use net_send to check if simultaneous spike occurred (otherwise bug when using mpi)
                }
                if ((RLon == 1) && (-interval <= RLwindanti)) { tlastantielig = t } : If RL and anti-Hebbian eligibility traces are turned on, and the interval falls within the maximum window for eligibility, remember the eligibilty trace start at the current time.
            }
            tlastpre = t : Remember the current spike time for next NET_RECEIVE.  

        : Else, if we receive a negative weight value, we are receiving a post-synaptic spike (and thus need to check for a Hebbian event, since the pre-synaptic weight must be earlier).    
        } else {            
            interval = t - tlastpre : Get the interval; interval is positive
            if  ((tlastpre > -1) && (interval > 1.0)) { : If we had a pre-synaptic spike and a non-zero interval...
                if (STDPon == 1) { : If STDP learning is turned on...
                    if (verbose > 0) {printf("net_send(1,-1)\n")}
                    net_send(1,-1) : instead of updating weight directly, use net_send to check if simultaneous spike occurred (otherwise bug when using mpi)
                }
                if ((RLon == 1) && (interval <= RLwindhebb)) { 
                    tlasthebbelig = t} : If RL and Hebbian eligibility traces are turned on, and the interval falls within the maximum window for eligibility, remember the eligibilty trace start at the current time.
            }
            tlastpost = t : Remember the current spike time for next NET_RECEIVE.
        }
    }
    if (verbose > 0)  { printf("t=%f (AFTER) tlaspre=%f, tlastpost=%f, flag=%f, w=%f, deltaw=%f \n",t,tlastpre, tlastpost,flag,w,deltaw) }
}

PROCEDURE reward_punish(reinf) {
    if (RLon == 1) { : If RL is turned on...
        deltaw = 0.0 : Start the weight change as being 0.
        deltaw = deltaw + reinf * hebbRL() : If we have the Hebbian eligibility traces on, add their effect in.   
        deltaw = deltaw + reinf * antiRL() : If we have the anti-Hebbian eligibility traces on, add their effect in.
        if (softthresh == 1) { deltaw = softthreshold(deltaw) }  : If we have soft-thresholding on, apply it.  
        adjustweight(deltaw) : Adjust the weight.
        if (verbose > 0) { printf("RL event: t = %f ms; reinf = %f; RLhebbwt = %f; RLlenhebb = %f; tlasthebbelig = %f; deltaw = %f\n",t,reinf,RLhebbwt,RLlenhebb,tlasthebbelig, deltaw) } : Show weight update information if debugging on.     
    }
}

FUNCTION hebbRL() {
    if ((RLon == 0) || (tlasthebbelig < 0.0)) { hebbRL = 0.0  } : If RL is turned off or eligibility has not occurred yet, return 0.0.
    else if (useRLexp == 0) { : If we are using a binary (i.e. square-wave) eligibility traces...
        if (t - tlasthebbelig <= RLlenhebb) { hebbRL = RLhebbwt } : If we are within the length of the eligibility trace...
        else { hebbRL = 0.0 } : Otherwise (outside the length), return 0.0.
    } 
    else { hebbRL = RLhebbwt * exp((tlasthebbelig - t) / RLlenhebb) } : Otherwise (if we're using an exponential decay traces)...use the Hebbian decay to calculate the gain.

}

FUNCTION antiRL() {
    if ((RLon == 0) || (tlastantielig < 0.0)) { antiRL = 0.0 } : If RL is turned off or eligibility has not occurred yet, return 0.0.
    else if (useRLexp == 0) { : If we are using a binary (i.e. square-wave) eligibility traces...
        if (t - tlastantielig <= RLlenanti) { antiRL = RLantiwt } : If we are within the length of the eligibility trace...
        else {antiRL = 0.0 } : Otherwise (outside the length), return 0.0.
    }
    else { antiRL = RLantiwt * exp((tlastantielig - t) / RLlenanti) } : Otherwise (if we're using an exponential decay traces), use the anti-Hebbian decay to calculate the gain.  
}

FUNCTION softthreshold(rawwc) {
    if (rawwc >= 0) { softthreshold = rawwc * (1.0 - synweight / wmax) } : If the weight change is non-negative, scale by 1 - weight / wmax.
    else { softthreshold = rawwc * synweight / wmax } : Otherwise (the weight change is negative), scale by weight / wmax.    
}

PROCEDURE adjustweight(wc) {
   synweight = synweight + wc : apply the synaptic modification, and then clip the weight if necessary to make sure it's between 0 and wmax.
   if (synweight > wmax) { synweight = wmax }
   if (synweight < 0) { synweight = 0 }
}

iomaganaris commented 2 years ago

Hello @jhielson ,

I have edited the mod files of the Izhikevich model to be compatible with CoreNEURON here: https://github.com/iomaganaris/39948/commit/aba2e4a5a5eaf5f7f8ef2ee9e5642df7b6feb578 For stdp.mod the situation is a bit more complex since you need to follow the guide in https://github.com/BlueBrain/CoreNeuron/blob/master/docs/userdoc/MemoryManagement/bbcorepointer.md to transform the POINTER variable to BBCOREPOINTER so that it can be transferred from NEURON to CoreNEURON.

jhielson commented 2 years ago

Dear @iomaganaris ,

I was using 'sudo' before. With my own user I get the following error:

VERBOSE=1 nrnivmodl -coreneuron .
/home/jhielson/catkin_ws/src/cerebellum/src
Mod files: "./izhi2007a.mod" "./izhi2007bS.mod" "./spikeout.mod" "./SynExp2NMDA.mod"

COBJS=''
[INFO] Running: make -j4 -f /home/jhielson/nrn-install/share/coreneuron/nrnivmodl_core_makefile ROOT=/home/jhielson/nrn-install MODS_PATH=x86_64/corenrn/mod2c BUILD_TYPE=STATIC
Default NMODL flags: 
/usr/bin/perl /home/jhielson/nrn-install/share/coreneuron/mod_func.c.pl SynExp2NMDA.mod exp2syn.mod expsyn.mod hh.mod izhi2007a.mod izhi2007bS.mod netstim.mod passive.mod pattern.mod spikeout.mod stim.mod svclmp.mod > x86_64/corenrn/mod2c/_mod_func.cpp.tmp
cd x86_64/corenrn/mod2c; \
/usr/bin/python3.6 /home/jhielson/nrn-install/share/coreneuron/kinderiv.py;
diff -q x86_64/corenrn/mod2c/_mod_func.cpp.tmp x86_64/corenrn/mod2c/_mod_func.cpp || \
mv x86_64/corenrn/mod2c/_mod_func.cpp.tmp x86_64/corenrn/mod2c/_mod_func.cpp
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc++ -O3 -g -acc -gpu=cuda11.3,cc60,cc70 --diag_suppress 161,177,550 -mp   -g  -O2  --c++14 -fPIC  -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DCORENEURON_BUILD -DHAVE_MALLOC_H -DEIGEN_DONT_VECTORIZE=1 -DNRNMPI=1 -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DLAYOUT=0 -DDISABLE_HOC_EXP -DENABLE_SPLAYTREE_QUEUING  -I/home/jhielson/nrn-install/include -I/home/jhielson/nrn-install/include/coreneuron/utils/randoms  -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/include  -c x86_64/corenrn/mod2c/izhi2007a.cpp -o x86_64/corenrn/build/izhi2007a.o
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc++ -O3 -g -acc -gpu=cuda11.3,cc60,cc70 --diag_suppress 161,177,550 -mp   -g  -O2  --c++14 -fPIC  -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DCORENEURON_BUILD -DHAVE_MALLOC_H -DEIGEN_DONT_VECTORIZE=1 -DNRNMPI=1 -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DLAYOUT=0 -DDISABLE_HOC_EXP -DENABLE_SPLAYTREE_QUEUING  -I/home/jhielson/nrn-install/include -I/home/jhielson/nrn-install/include/coreneuron/utils/randoms  -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/include  -c x86_64/corenrn/mod2c/izhi2007bS.cpp -o x86_64/corenrn/build/izhi2007bS.o
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc++ -O3 -g -acc -gpu=cuda11.3,cc60,cc70 --diag_suppress 161,177,550 -mp   -g  -O2  --c++14 -fPIC  -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DCORENEURON_BUILD -DHAVE_MALLOC_H -DEIGEN_DONT_VECTORIZE=1 -DNRNMPI=1 -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DLAYOUT=0 -DDISABLE_HOC_EXP -DENABLE_SPLAYTREE_QUEUING  -I/home/jhielson/nrn-install/include -I/home/jhielson/nrn-install/include/coreneuron/utils/randoms  -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/include  -c x86_64/corenrn/mod2c/spikeout.cpp -o x86_64/corenrn/build/spikeout.o
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc++ -O3 -g -acc -gpu=cuda11.3,cc60,cc70 --diag_suppress 161,177,550 -mp   -g  -O2  --c++14 -fPIC  -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DCORENEURON_BUILD -DHAVE_MALLOC_H -DEIGEN_DONT_VECTORIZE=1 -DNRNMPI=1 -DMPI_NO_CPPBIND=1 -DOMPI_SKIP_MPICXX=1 -DMPICH_SKIP_MPICXX=1 -DLAYOUT=0 -DDISABLE_HOC_EXP -DENABLE_SPLAYTREE_QUEUING  -I/home/jhielson/nrn-install/include -I/home/jhielson/nrn-install/include/coreneuron/utils/randoms  -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/include  -c x86_64/corenrn/mod2c/stim.cpp -o x86_64/corenrn/build/stim.o
NVC++-S-1061-Procedures called in a compute region must have acc routine information - fopen (x86_64/corenrn/mod2c/izhi2007a.cpp: 627)
NVC++-S-1000-Call in OpenACC region to procedure '_ZN10coreneuron13hoc_execerrorEPKcS1_' which has no acc routine information (x86_64/corenrn/mod2c/izhi2007bS.cpp: 406)
NVC++/x86-64 Linux 21.5-0: compilation completed with severe errors
/home/jhielson/nrn-install/share/coreneuron/nrnivmodl_core_makefile:213: recipe for target 'x86_64/corenrn/build/izhi2007bS.o' failed
make: *** [x86_64/corenrn/build/izhi2007bS.o] Error 2
make: *** Waiting for unfinished jobs....
NVC++/x86-64 Linux 21.5-0: compilation completed with severe errors
/home/jhielson/nrn-install/share/coreneuron/nrnivmodl_core_makefile:213: recipe for target 'x86_64/corenrn/build/izhi2007a.o' failed
make: *** [x86_64/corenrn/build/izhi2007a.o] Error 2
NVC++-W-1053-External and Static variables are not supported in acc routine - _ZN10coreneuron11nrn_threadsE (x86_64/corenrn/mod2c/spikeout.cpp: 422)
NVC++/x86-64 Linux 21.5-0: compilation completed with warnings

jhielson commented 2 years ago

Hi @iomaganaris ,

It seems to be a bit similar to the following issue: https://github.com/BlueBrain/CoreNeuron/issues/580#issuecomment-869745985

But now with acc routines:

NVC++-S-1061-Procedures called in a compute region must have acc routine information - fopen (x86_64/corenrn/mod2c/izhi2007a.cpp: 627)
NVC++-S-1000-Call in OpenACC region to procedure '_ZN10coreneuron13hoc_execerrorEPKcS1_' which has no acc routine information (x86_64/corenrn/mod2c/izhi2007bS.cpp: 406)

iomaganaris commented 2 years ago

Hello @jhielson ,

I have updated the mod files to make them compatible with GPU compilation in https://github.com/iomaganaris/39948/commit/e5187a9c13f3587b44e1ffdff8cd64a37e99173e. Please try with these changes and let me know if it works.

jhielson commented 2 years ago

Hi @iomaganaris ,

Thanks! It now compiles properly but when I use one of them in my code it generates an error.

Compilation of the mod file:

VERBOSE=0 nrnivmodl -coreneuron .
/home/jhielson/catkin_ws/src/cerebellum/src
Mod files: "./izhi2007a.mod" "./izhi2007bS.mod" "./spikeout.mod" "./SynExp2NMDA.mod"

COBJS=''
[INFO] Running: make -j4 -f /home/jhielson/nrn-install/share/coreneuron/nrnivmodl_core_makefile ROOT=/home/jhielson/nrn-install MODS_PATH=x86_64/corenrn/mod2c BUILD_TYPE=STATIC
Default NMODL flags: 
/usr/bin/perl /home/jhielson/nrn-install/share/coreneuron/mod_func.c.pl SynExp2NMDA.mod exp2syn.mod expsyn.mod hh.mod izhi2007a.mod izhi2007bS.mod netstim.mod passive.mod pattern.mod spikeout.mod stim.mod svclmp.mod > x86_64/corenrn/mod2c/_mod_func.cpp.tmp
cd x86_64/corenrn/mod2c; \
/usr/bin/python3.6 /home/jhielson/nrn-install/share/coreneuron/kinderiv.py;
diff -q x86_64/corenrn/mod2c/_mod_func.cpp.tmp x86_64/corenrn/mod2c/_mod_func.cpp || \
mv x86_64/corenrn/mod2c/_mod_func.cpp.tmp x86_64/corenrn/mod2c/_mod_func.cpp
mkdir -p x86_64/corenrn/build/scopmath; \
cd x86_64/corenrn/build/scopmath && ar -x /home/jhielson/nrn-install/lib/libscopmath.a && cd -;\
rm -f x86_64/libcorenrnmech.a; \
ar cq x86_64/libcorenrnmech.a x86_64/corenrn/build/enginemech.o x86_64/corenrn/build/_mod_func.o x86_64/corenrn/build/_dimplic.o x86_64/corenrn/build/SynExp2NMDA.o x86_64/corenrn/build/exp2syn.o x86_64/corenrn/build/expsyn.o x86_64/corenrn/build/hh.o x86_64/corenrn/build/izhi2007a.o x86_64/corenrn/build/izhi2007bS.o x86_64/corenrn/build/netstim.o x86_64/corenrn/build/passive.o x86_64/corenrn/build/pattern.o x86_64/corenrn/build/spikeout.o x86_64/corenrn/build/stim.o x86_64/corenrn/build/svclmp.o  x86_64/corenrn/build/scopmath/*.o;
/home/jhielson/catkin_ws/src/cerebellum/src
rm -rf x86_64/.libs/libcorenrnmech.a; \
mkdir -p x86_64/.libs; \
ln -s ../libcorenrnmech.a x86_64/.libs/libcorenrnmech.a
 => Binary creating x86_64/special-core
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc++ -O3 -g -acc -gpu=cuda11.3,cc60,cc70 --diag_suppress 161,177,550 -mp   -g  -O2  --c++14  -o x86_64/special-core /home/jhielson/nrn-install/share/coreneuron/coreneuron.cpp \
  -I/home/jhielson/nrn-install/include  \
  -Lx86_64 -lcorenrnmech -L/home/jhielson/nrn-install/lib -lcoreneuron       -lcudacoreneuron /usr/local/cuda-11.3/lib64/libcudart_static.a -ldl /usr/lib/x86_64-linux-gnu/librt.so /opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi.so \
  -Wl,-rpath,'$ORIGIN' -Wl,-rpath,/home/jhielson/nrn-install/lib
/home/jhielson/nrn-install/share/coreneuron/coreneuron.cpp:
[INFO] MOD files built successfully for CoreNEURON
 -> Compiling mod_func.cpp
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc -g  -O2  -O3 -g  -I.   -I/home/jhielson/nrn-install/include  -I/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/include -fPIC -c mod_func.cpp -o mod_func.o
 => LINKING shared library ./libnrnmech.so
/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/compilers/bin/nvc++ -g  -O2  -O3 -g --c++11 -shared -fPIC  -I /home/jhielson/nrn-install/include -o ./libnrnmech.so -Wl,-soname,libnrnmech.so \
  ./mod_func.o ./izhi2007a.o ./izhi2007bS.o ./spikeout.o ./SynExp2NMDA.o  -L/home/jhielson/nrn-install/lib -lnrniv -Wl,-rpath,/home/jhielson/nrn-install/lib    /usr/lib/x86_64-linux-gnu/libreadline.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/libcurses.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/libform.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/libpython3.6m.so -Wl,-rpath,/usr/lib/x86_64-linux-gnu /opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/lib/libmpi.so -Wl,-rpath,/opt/nvidia/hpc_sdk/Linux_x86_64/21.5/comm_libs/openmpi/openmpi-3.1.5/lib
rm -f ./.libs/libnrnmech.so ; mkdir -p ./.libs ; cp ./libnrnmech.so ./.libs/libnrnmech.so
Successfully created x86_64/special

My simulation output:

mpirun -n 6 ./x86_64/special -mpi -python completeModel.py 
numprocs=6
NEURON -- VERSION 8.0a-590-ge40208d8+ master (e40208d8+) 2021-06-18
Duke, Yale, and the BlueBrain Project -- Copyright 1984-2021
See http://neuron.yale.edu/neuron/credits

Additional mechanisms from files
 "./izhi2007a.mod" "./izhi2007bS.mod" "./spikeout.mod" "./SynExp2NMDA.mod"

Start time:  2022-02-03 12:39:09.130380

Creating network of 117 cell populations on 6 hosts...
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
ERROR: Some mechanisms and/or ions were not inserted (for details run with cfg.verbose=True). Make sure the required mod files are compiled.
>>> >>> >>> >>> >>> >>> Traceback (most recent call last):
  File "completeModel.py", line 1943, in <module>
    output = network.simulate()
  File "completeModel.py", line 1569, in simulate
    sim.net.createCells()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/network.py", line 86, in createCells
    newCells = ipop.createCells() # create cells for this pop using Pop method
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 79, in createCells
    cells = self.createCellsFixedNum()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 175, in createCellsFixedNum
    cells.append(self.cellModelClass(gid, cellTags)) # instantiate Cell object
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 52, in __init__
    if create: self.create()  # create cell
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 107, in create
    self.createNEURONObj(prop)  # add sections, mechanisms, synaptic mechanisms, geometry and topolgy specified by this property set
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 432, in createNEURONObj
    setattr(sec['pointps'][pointpName]['hObj'], pointpParamName, pointpParamValue)
LookupError: 'f' is not a defined hoc variable name.

Traceback (most recent call last):
  File "completeModel.py", line 1943, in <module>
    output = network.simulate()
  File "completeModel.py", line 1569, in simulate
    sim.net.createCells()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/network.py", line 86, in createCells
    newCells = ipop.createCells() # create cells for this pop using Pop method
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 79, in createCells
    cells = self.createCellsFixedNum()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 175, in createCellsFixedNum
    cells.append(self.cellModelClass(gid, cellTags)) # instantiate Cell object
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 52, in __init__
    if create: self.create()  # create cell
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 107, in create
    self.createNEURONObj(prop)  # add sections, mechanisms, synaptic mechanisms, geometry and topolgy specified by this property set
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 432, in createNEURONObj
    setattr(sec['pointps'][pointpName]['hObj'], pointpParamName, pointpParamValue)
LookupError: 'f' is not a defined hoc variable name.

Traceback (most recent call last):
  File "completeModel.py", line 1943, in <module>
    output = network.simulate()
  File "completeModel.py", line 1569, in simulate
    sim.net.createCells()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/network.py", line 86, in createCells
    newCells = ipop.createCells() # create cells for this pop using Pop method
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 79, in createCells
    cells = self.createCellsFixedNum()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 175, in createCellsFixedNum
    cells.append(self.cellModelClass(gid, cellTags)) # instantiate Cell object
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 52, in __init__
    if create: self.create()  # create cell
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 107, in create
    self.createNEURONObj(prop)  # add sections, mechanisms, synaptic mechanisms, geometry and topolgy specified by this property set
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 432, in createNEURONObj
    setattr(sec['pointps'][pointpName]['hObj'], pointpParamName, pointpParamValue)
LookupError: 'f' is not a defined hoc variable name.

Traceback (most recent call last):
  File "completeModel.py", line 1943, in <module>
    output = network.simulate()
  File "completeModel.py", line 1569, in simulate
    sim.net.createCells()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/network.py", line 86, in createCells
    newCells = ipop.createCells() # create cells for this pop using Pop method
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 79, in createCells
    cells = self.createCellsFixedNum()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 175, in createCellsFixedNum
    cells.append(self.cellModelClass(gid, cellTags)) # instantiate Cell object
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 52, in __init__
    if create: self.create()  # create cell
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 107, in create
    self.createNEURONObj(prop)  # add sections, mechanisms, synaptic mechanisms, geometry and topolgy specified by this property set
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 432, in createNEURONObj
    setattr(sec['pointps'][pointpName]['hObj'], pointpParamName, pointpParamValue)
LookupError: 'f' is not a defined hoc variable name.

Traceback (most recent call last):
  File "completeModel.py", line 1943, in <module>
    output = network.simulate()
  File "completeModel.py", line 1569, in simulate
    sim.net.createCells()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/network.py", line 86, in createCells
    newCells = ipop.createCells() # create cells for this pop using Pop method
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 79, in createCells
    cells = self.createCellsFixedNum()
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/network/pop.py", line 175, in createCellsFixedNum
    cells.append(self.cellModelClass(gid, cellTags)) # instantiate Cell object
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 52, in __init__
    if create: self.create()  # create cell
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 107, in create
    self.createNEURONObj(prop)  # add sections, mechanisms, synaptic mechanisms, geometry and topolgy specified by this property set
  File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 432, in createNEURONObj
    setattr(sec['pointps'][pointpName]['hObj'], pointpParamName, pointpParamValue)
LookupError: 'f' is not a defined hoc variable name.

iomaganaris commented 2 years ago

Hello @jhielson ,

It seems to me like something is not properly set in the simulation you're trying to run since I see that the mechanisms from the mod files are loaded in NEURON:

Additional mechanisms from files
 "./izhi2007a.mod" "./izhi2007bS.mod" "./spikeout.mod" "./SynExp2NMDA.mod"

The problem seems to be coming from:

 File "/home/jhielson/.local/lib/python3.6/site-packages/netpyne/cell/compartCell.py", line 432, in createNEURONObj
    setattr(sec['pointps'][pointpName]['hObj'], pointpParamName, pointpParamValue)
LookupError: 'f' is not a defined hoc variable name.

https://github.com/suny-downstate-medical-center/netpyne/blob/eedb155a7467c4a1deeed1e50eb5db4ea46db783/netpyne/cell/compartCell.py#L432 Maybe something is not set properly in the circuit configuration?

jhielson commented 2 years ago

Hi @iomaganaris ,

You are right. I was using an old setup in my code. Thank you!

        cellRule['secs']['soma']['pointps']['Izhi'] = {'mod': 'Izhi2007bS',
                                                       'a': 0.1,
                                                       'b': 0.2,
                                                       'c': -65,
                                                       'd': 2,
                                                       'f': 5,
                                                       'g': 140,
                                                       'thresh': 30}

jhielson commented 2 years ago

Hi @iomaganaris,

Can you help me with this new issue? I am getting the following error after running my code:

Error creating point process mechanism STN in cell with gid 815
5 ./x86_64/special: NetCon[0] source is missing
5  near line 0
5  objref hoc_obj_[2]

Output:

$ mpirun -n 6 ./x86_64/special -mpi -python completeModel.py 
numprocs=6
NEURON -- VERSION 8.0a-590-ge40208d8+ master (e40208d8+) 2021-06-18
Duke, Yale, and the BlueBrain Project -- Copyright 1984-2021
See http://neuron.yale.edu/neuron/credits

Additional mechanisms from files
 "./GP.mod" "./izhi2007a.mod" "./izhi2007bS.mod" "./spikeout.mod" "./STN.mod" "./Str.mod" "./SynExp2NMDA.mod" "./thalamus.mod"

Start time:  2022-02-10 13:51:53.086230

Creating network of 668 cell populations on 6 hosts...
Error creating point process mechanism STN in cell with gid 815
5 ./x86_64/special: NetCon[0] source is missing
5  near line 0
5  objref hoc_obj_[2]
                   ^
Error creating point process mechanism STN in cell with gid 812
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 5 in communicator MPI_COMM_WORLD
with errorcode -1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------

That's how I designed a STN neuron:

    def stnCellRules(self, gkcabar=3.8e-3):
        cellRule = {'conds': {'cellModel': 'STN', 'cellType': 'STN'}, 'secs': {}}
        cellRule['secs']['soma'] = {'geom': {}, 'mechs': {}}
        cellRule['secs']['soma']['geom'] = {'diam': 5.642,
                                            'L': 5.642,
                                            'Ra': 1,
                                            'nseg': 1}
        cellRule['secs']['soma']['mechs']['STN'] = {'dbs': self.dbs,'gkcabar': gkcabar} 
        cellRule['secs']['soma']['vinit'] = random.gauss(-62, 5)
        cellRule['secs']['soma']['threshold'] = -10
        self.netParams.cellParams['STN'] = cellRule

That's the STN.mod file:

TITLE  STN ion channels for single compartment model

COMMENT
//program from Basal ganglia network model of subthalamic deep brain stimulation (Hahn and McIntyre 2010)
//derived from: Otsuka model
//available on : 
//https://senselab.med.yale.edu/modeldb/ShowModel.cshtml?model=127388
//Author: Daniel Tamashiro, Unicamp 
//Updated 08/16/2016
ENDCOMMENT

UNITSON

NEURON {
    SUFFIX STN
    NONSPECIFIC_CURRENT ilk
    NONSPECIFIC_CURRENT icaT,icaL,ikD,ikA,ikAHP,ina, idbs

    RANGE gnabar, ena, m_inf, h_inf, tau_h, tau_m          : fast sodium
    RANGE gkdrbar, ek, n_inf, tau_n                   : delayed K rectifier
    RANGE gl, el                                     : leak
    RANGE gcatbar, p_inf, tau_p, q_inf, tau_q          : T-type ca current
    RANGE gcalbar, eca, c_inf, d1_inf, d2_inf, tau_c, tau_d1, tau_d2  : L-type ca current
    RANGE gkabar, ek, a_inf, tau_a, b_inf, tau_b     : A-type K current
    RANGE gkcabar, ek, r_inf                       : ca dependent AHP K current

    RANGE periodo, tezao, cai, dbs
}

UNITS {
    (mA) = (milliamp)
    (mV) = (millivolt)
    (S)  = (siemens)
    (molar) = (1/liter)
    (mM)    = (millimolar)
}

PARAMETER {
    ena = 60    (mV)
    ek = -90    (mV)
    cao = 2000  
    con =12.8392 (mV)

    tezao = 7.6923 (ms)
    dbs

:Fast Na channel
    gnabar   = 49e-3 (S/cm2) 
    theta_m = -40 (mV)
    theta_h = -45.5 (mV) 
    k_m = -8 (mV)   
    k_h = 6.4 (mV)   
    tau_m0 = 0.2 (ms)
    tau_m1 = 3 (ms)
    tau_h0 = 0 (ms)
    tau_h1 = 24.5 (ms) 
    tht_m = -53 (mV)
    tht_h1 = -50 (mV)
    tht_h2 = -50 (mV)
    sig_m = -0.7 (mV)
    sig_h1 = -15 (mV)
    sig_h2 = 16 (mV)

: Delayed rectifier K
    gkdrbar  = 57e-3    (S/cm2)  
    theta_n = -41 (mV)
    k_n = -14 (mV)     
    tau_n0 = 0 (ms)
    tau_n1 = 11 (ms) 
    tht_n1 = -40 (mV)
    tht_n2 = -40 (mV)
    sig_n1 = -40 (mV)
    sig_n2 = 50 (mV) 

:Leakage current
    gl  = 0.35e-3   (S/cm2)
    el  = -60   (mV)

:T-type ca current
    gcatbar   = 5e-3 (S/cm2)  
    theta_p = -56 (mV)
    theta_q = -85 (mV) 
    k_p = -6.7 (mV)    
    k_q = 5.8 (mV)  
    tau_p0 = 5 (ms)
    tau_p1 = 0.33 (ms)
    tau_q0 = 0 (ms)
    tau_q1 = 400 (ms) 
    tht_p1 = -27 (mV)
    tht_p2 = -102 (mV)
    tht_q1 = -50 (mV)
    tht_q2 = -50 (mV)
    sig_p1 = -10 (mV)
    sig_p2 = 15 (mV)    
    sig_q1 = -15 (mV)
    sig_q2 = 16 (mV)    

:Ca L current
    gcalbar   = 15e-3 (S/cm2) 
    theta_c = -30.6 (mV)
    theta_d1 = -60 (mV)
    theta_d2 = 0.1 (mV)
    k_c = -5 (mV)   
    k_d1 = 7.5 (mV)
    k_d2 = 0.02 (mV)
    tau_c0 = 45 (ms)
    tau_c1 = 10 (ms)
    tau_d10 = 400 (ms)
    tau_d11 = 500 (ms)
    tht_c1 = -27 (mV)
    tht_c2 = -50 (mV)
    tht_d11 = -40 (mV)
    tht_d12 = -20 (mV)
    sig_c1 = -20 (mV)
    sig_c2 = 15 (mV)    
    sig_d11 = -15 (mV)
    sig_d12 = 20 (mV)   
    tau_d2 = 130 (ms)

:A current
    gkabar  = 5e-3  (S/cm2)  
    theta_a = -45 (mV)  
    theta_b = -90 (mV) 
    k_a = -14.7 (mV)    
    k_b = 7.5 (mV)      
    tau_a0 = 1 (ms)
    tau_a1 = 1 (ms)
    tau_b0 = 0 (ms)
    tau_b1 = 200 (ms) 
    tht_a = -40 (mV)
    tht_b1 = -60 (mV)
    tht_b2 = -40 (mV)
    sig_a = -0.5 (mV)
    sig_b1 = -30 (mV)
    sig_b2 = 10 (mV)    

:AHP current (Ca dependent K current)
    gkcabar   = 1e-3 (S/cm2) 
    theta_r = 0.17 (mV) 
    k_r = -0.08 (mV) 
    tau_r = 2 (ms)

:cai 
    acai=5.18e-3 (cm2/mA/ms)
    bcai=2e-3 (1/ms) 
}

ASSIGNED {
    v   (mV)
    ina (mA/cm2)
    ikD (mA/cm2)   
    ikA (mA/cm2) 
    ikAHP(mA/cm2)  
    icaT(mA/cm2) 
    icaL (mA/cm2)
    ilk (mA/cm2)
    idbs (mA/cm2)

    periodo (ms)

:Fast Na
    h_inf
    tau_h   (ms)
    m_inf
    tau_m   (ms) 

:Delayed rectifier
    n_inf
    tau_n   (ms)

:ca T current
    p_inf
    q_inf
    tau_p   (ms)
    tau_q   (ms)
    eca     (mV)   :calc from Nernst

:ca L current
    c_inf
    tau_c   (ms)
    d1_inf
    tau_d1  (ms)
    d2_inf

:A current
    a_inf
    tau_a   (ms)
    b_inf
    tau_b   (ms)

:AHP (Ca dependent K current)
    r_inf
}

STATE {
    m h n
    p q 
    c d1 d2
    cai 
    a b r
}

BREAKPOINT {
    SOLVE states METHOD cnexp

    eca = con*log(cao/cai)

    ina   = gnabar * m*m*m*h * (v - ena)
    ikD   = gkdrbar * n^4 * (v - ek)

    ikA   = gkabar * a*a*b * (v - ek)
    ikAHP   = gkcabar *r*r* (v - ek)

    icaT   = gcatbar * p*p*q * (v - eca)
    icaL   = gcalbar * c*c*d1*d2 * (v - eca)

    ilk = gl * (v - el)

    if (t >= periodo + tezao){
        periodo = periodo + tezao
    }
    if (t >= periodo && t <= periodo + 0.3) { 
        idbs = -0.3*dbs
    } else{
        idbs = 0
    }
}

DERIVATIVE states {   
    evaluate_fct(v)
    h' = (h_inf - h)/tau_h
    m' = (m_inf - m)/tau_m
    n' = (n_inf - n)/tau_n
    p' = (p_inf - p)/tau_p
    q' = (q_inf - q)/tau_q

    c' = (c_inf - c)/tau_c
    d1' = (d1_inf - d1)/tau_d1
    d2' = (d2_inf - d2)/tau_d2

    cai' =  -acai*(icaL+icaT)-bcai*cai

    a' = (a_inf - a)/tau_a
    b' = (b_inf - b)/tau_b

    r' = (r_inf - r)/tau_r
}

UNITSOFF

INITIAL {
    evaluate_fct(v)
    m = m_inf 
    h = h_inf   
    n = n_inf   
    p = p_inf 
    q = q_inf  

    c = c_inf 
    d1 = d1_inf  
    d2 = d2_inf   

    a = a_inf 
    b = b_inf   

    r = r_inf 
    cai= 0.005

    periodo = 0
}

PROCEDURE evaluate_fct(v(mV)) { 
:Fast Na current
    h_inf = 1/(1+exp((v-theta_h)/k_h))
    m_inf = 1/(1+exp((v-theta_m)/k_m))
    tau_h = tau_h0 + tau_h1/(exp(-(v-tht_h1)/sig_h1) + exp(-(v-tht_h2)/sig_h2)) 
    tau_m = tau_m0 + tau_m1/(1+exp(-(v-tht_m)/sig_m)) 

:Delayed rectifier K
    n_inf = 1/(1+exp((v-theta_n)/k_n))
    tau_n = tau_n0 + tau_n1/(exp(-(v-tht_n1)/sig_n1) + exp(-(v-tht_n2)/sig_n2)) 

:Ca T current
    p_inf = 1/(1+exp((v-theta_p)/k_p))
    q_inf = 1/(1+exp((v-theta_q)/k_q))
    tau_p = tau_p0 + tau_p1/(exp(-(v-tht_p1)/sig_p1) + exp(-(v-tht_p2)/sig_p2)) 
    tau_q = tau_q0 + tau_q1/(exp(-(v-tht_q1)/sig_q1) + exp(-(v-tht_q2)/sig_q2))

:Ca L current
    c_inf = 1/(1+exp((v-theta_c)/k_c))
    d1_inf = 1/(1+exp((v-theta_d1)/k_d1))
    d2_inf = 1/(1+exp((v-theta_d2)/k_d2))
    tau_c = tau_c0 + tau_c1/(exp(-(v-tht_c1)/sig_c1) + exp(-(v-tht_c2)/sig_c2))  
    tau_d1 = tau_d10 + tau_d11/(exp(-(v-tht_d11)/sig_d11) + exp(-(v-tht_d12)/sig_d12))  

:A current
    a_inf = 1/(1+exp((v-theta_a)/k_a))
    b_inf = 1/(1+exp((v-theta_b)/k_b))
    tau_a = tau_a0 + tau_a1/(1+exp(-(v-tht_a)/sig_a))
    tau_b = tau_b0 + tau_b1/(exp(-(v-tht_b1)/sig_b1) + exp(-(v-tht_b2)/sig_b2))  

:AHP current
    r_inf = 1/(1+exp((v-theta_r)/k_r))
}

UNITSON

I have tested the code without using Coreneuron and it works properly. I think the issue might be in the following line:

        cellRule['secs']['soma']['mechs']['STN'] = {'dbs': self.dbs,'gkcabar': gkcabar}

Maybe, I am missing something there.

jhielson commented 2 years ago

Dear @iomaganaris,

After a long debugging, we could find the solution for that issue. It seems the name of the cell mechanism can not be the same used for the cell type and cell model. After changing that, things started to work.

Before:

   def stnCellRules(self, gkcabar=3.8e-3):
        cellRule = {'conds': {'cellModel': 'STN', 'cellType': 'STN'}, 'secs': {}}
        cellRule['secs']['soma'] = {'geom': {}, 'mechs': {}}
        cellRule['secs']['soma']['geom'] = {'diam': 5.642,
                                            'L': 5.642,
                                            'Ra': 1,
                                            'nseg': 1}
        cellRule['secs']['soma']['mechs']['STN'] = {'dbs': self.dbs,'gkcabar': gkcabar} 
        cellRule['secs']['soma']['vinit'] = random.gauss(-62, 5)
        cellRule['secs']['soma']['threshold'] = -10
        self.netParams.cellParams['STN'] = cellRule

After:

        cellRule = {'conds': {'cellModel':'STN', 'cellType':'STN'}, 'secs': {}}
        cellRule['secs']['soma'] = {'geom': {}, 'mechs': {}}
        cellRule['secs']['soma']['geom'] = {'diam': 5.642,
                                            'L': 5.642,
                                            'Ra': 1,
                                            'nseg': 1}
        cellRule['secs']['soma']['mechs']['SubTN'] = {'dbs': self.dbs,'gkcabar': gkcabar} 
        cellRule['secs']['soma']['vinit'] = random.gauss(-62, 5)
        cellRule['secs']['soma']['threshold'] = -10
        self.netParams.cellParams['STN'] = cellRule

PS: I also changed its name inside the mod file.

suny-downstate-medical-center / netpyne

[documentation] Activating the coreNeuron package in a NetPyNE example #611