www.neuron.yale.edu

I think the easiest installation is to try to remove all of what you have done and download https://neuron.yale.edu/ftp/neuron/versions/alpha/nrn-7.8.0-82-g5104c2e7-osx-py-27-36-37-38.pkg and double click on it. If that works and you then want to build from github sources, I can give the recipe for ...

You didn't mention whether you installed from sources or a pkg file? I'll assume the former. You didn't mention how you configured. So I don't know where the neuron module was installed. Or where the bulk of neuron itself was installed. With your PATH=$HOME/local/bin I assume you used cmake (configu...

The symbol that is not found would be in the interviews library and it is not clear to me why that is needed to successfully load the libnrnmech.so file. Can you try the following test which loads the mod files for the neurondemo python3 from neuron import h h.nrn_load_dll("/Applications/NEURON-7.7/...

Check the output of otool -L /Applications/NEURON-7.7/nrn/lib/python/neuron/hoc.so to see if it cannot find any of the libraries that library is trying to load. It may be that you need to install/upgrade your command line tools and XQuartz. The latter comes from xquartz.org. For the former, when I t...

If you start a terminal and launch python or python3 what is the result of
import sys
sys.version

ld: library not found for -lX11

It appears that XQuartz is not installed. You can do that from xquartz.org

The problem with VecStim is that the mod file implementation did not envision this style of usage. I would recommend adding a restart procedure which, after you replace the Vector in the middle of a simulation, would set the counter to the first element of the Vector and do a net_send to activate th...

Yes. It is not enough any longer that the package be signed, it must also be notarized (now a $100 annual fee). Notarization means that Apple scanned the app for known malicious software. From https://support.apple.com/en-us/HT202491 How to open an app that hasn’t been notarized or is from an uniden...

MODLUNIT and NEURONHOME should be export MODLUNIT=/home/ziemek/usr/local/nrn/share/nrn/lib/nrnunits.lib export NEURONHOME=/home/ziemek/usr/local/nrn/share/nrn (I'm not sure but if you have NEURONHOME set, you may not need MODLUNIT) If you want to launch python and import neuron then export PYTHONPAT...

I had forgotten about that. It was the default for SCoP. The fix on our part will be to replace the existing adrunge implementation with a thread safe version. Nevertheless, adrunge generally makes little sense in a NEURON context since the extra accuracy does not translate into a more accurate simu...

You did not mention whether the 7.7 version hangs on your Macbook pro and on comet. Or if NEURON 7.5 has no problem on Manjaro linux. There have been a lot of mpi changes since 7.5. If it does not take too long to run an example, you can send me a zip file with sufficient code to exhibit the problem...

So I assume that I can just add libs by hand to the PYTHONPATH, but could you tell me where I can find export PYTHONPATH=/where/you/installed/nrn/lib #often /usr/local/nrn/lib The default build installs the NEURON module in <prefix>/lib/python in order to be python independent of the python version...

On my machine (macOS Catalina Version10.15 Beta (19A573a)) (my most recent build in /Applications/NEURON-7.7 is NEURON -- VERSION 7.7.1-38-g1f3d46d9+ master (1f3d46d9+) 2019-09-13 ) I'm seeing for a typical libnrnmech.so created by nrnivmodl $ otool -L x86_64/.libs/libnrnmech.so ... /usr/lib/libncur...

An old way of getting the PARAMETER names of a mechanism is from neuron import h ms = h.MechanismStandard("hh", 1) parm_name = h.ref("") for i in range(ms.count()): ms.name(parm_name, i) print(parm_name[0]) https://www.neuron.yale.edu/neuron/static/py_doc/programming/mechstan.html#MechanismStandard....

Segmentation Fault errors under MPI with gap junctions when I had h.cvode.cache_efficient(1) I'd like to attempt to fix this bug. Can you send me a (hopefully small) version of your model that exhibits this error. (Michael.hines@yale.edu) An unrelated point is that it is now possible to avoid the "...