Hi Cristiano, Thanks for writing in. Of course, in the ideal case, you would not have to modify the generated C++ code as everything should be already contained within the NESTML model itself. So if there's a particular way in which NESTML is incomplete or does not allow you to express something, please also get back to us about this, we'd be glad to help. Of course in practice, it might very well occur during development that you want to modify the generated code just a little, for example to insert some printf() calls. The way to do this depends a little on how you are using NESTML. If you are using the API, with to_nest() and install_nest() and finally nest.Install(), the trick is to skip the install_nest() step, and run the compilation by hand. In the directory where the C++ code is generated, run "cmake -Dwith-nest=/path/to/your/nest-sim/install/bin/nest-config" and then "make install". The module can then be loaded again using nest.Install() in your Python script. To prevent from continuously having to modify generated code, you could also potentially insert your changes into the templates. These you can find in pynestml/codegeneration/resources_nest/{NeuronHeader.jinja2, NeuronClass.jinja2}. As for how the code works, I'm not sure if there's an extended description anywhere. The gist of it is that incoming events (like spikes) are handled in the handle() method, and time stepping is handled in update(). Hopefully the rest should speak mostly for itself, please do not hesitate to get in touch if you have any specific questions or run into any issues. Cheers, kind regards, Charl On Thu, Dec 10, 2020, at 18:44, Cristiano Alessandro wrote:

Hi! First of all, I agree with everything that Charl said. We are currently in the process of updating, extending and cleaning up our developer level documentation and I'm afraid it does not currently live up to our standards. That said, I have attached a tutorial talk I gave some years ago on the topic of developing neuron and synapse models by hand. Although some of the functions have been renamed since, it should still give you a good overview of the concepts and most important API functions. Moreover, you should be able to run `make doc` or `make fulldoc` in the NEST build directory to get the Doxygen documentation built. That will contain more documentation about the classes you are asking about. Cheers, Jochen! On 11.12.20 10:36, Cristiano Alessandro wrote: Dear Charl, Thank you so much for your kind reply. Could you elaborate a little more regarding the fact that incoming events (like spikes) are handled in the handle() method, and time stepping is handled in update()? When are these functions called and at what frequency? My overall understanding was that spikes are buffered in a RingBuffer, and that the function update() runs at the end of each time step, when the buffer is full (so that I can read all the events in the buffer)? What about the handle function? Would it be possible to know the methods and attributes of the following classes? nest::RingBuffer nest::SpikeEvent nest::Time Thank you again in advance for your help Best Cristiano On Fri, Dec 11, 2020 at 10:25 AM Charl Linssen <nest-users@turingbirds.com<mailto:nest-users@turingbirds.com>> wrote: Hi Cristiano, Thanks for writing in. Of course, in the ideal case, you would not have to modify the generated C++ code as everything should be already contained within the NESTML model itself. So if there's a particular way in which NESTML is incomplete or does not allow you to express something, please also get back to us about this, we'd be glad to help. Of course in practice, it might very well occur during development that you want to modify the generated code just a little, for example to insert some printf() calls. The way to do this depends a little on how you are using NESTML. If you are using the API, with to_nest() and install_nest() and finally nest.Install(), the trick is to skip the install_nest() step, and run the compilation by hand. In the directory where the C++ code is generated, run "cmake -Dwith-nest=/path/to/your/nest-sim/install/bin/nest-config" and then "make install". The module can then be loaded again using nest.Install() in your Python script. To prevent from continuously having to modify generated code, you could also potentially insert your changes into the templates. These you can find in pynestml/codegeneration/resources_nest/{NeuronHeader.jinja2, NeuronClass.jinja2}. As for how the code works, I'm not sure if there's an extended description anywhere. The gist of it is that incoming events (like spikes) are handled in the handle() method, and time stepping is handled in update(). Hopefully the rest should speak mostly for itself, please do not hesitate to get in touch if you have any specific questions or run into any issues. Cheers, kind regards, Charl On Thu, Dec 10, 2020, at 18:44, Cristiano Alessandro wrote: Dear all, I am new to developing NEST neuron models. I am aware I should use NEST-ML and I started that way, but I now need to modify the (successfully) generated C++ code. To do so, I would like to understand the details of the C++ implementation, including the overall logic of the class NEST-ML created (what is the functionality of the methods and attributes). Where can I find documentation about this? I would also like to take a look at the description of the NEST C++ developers library. I could not find these aspects in the Nest Developer Space. Thank you in advance. Best Cristiano _______________________________________________ NEST Users mailing list -- users@nest-simulator.org<mailto:users@nest-simulator.org> To unsubscribe send an email to users-leave@nest-simulator.org<mailto:users-leave@nest-simulator.org> _______________________________________________ NEST Users mailing list -- users@nest-simulator.org<mailto:users@nest-simulator.org> To unsubscribe send an email to users-leave@nest-simulator.org<mailto:users-leave@nest-simulator.org> _______________________________________________ NEST Users mailing list -- users@nest-simulator.org<mailto:users@nest-simulator.org> To unsubscribe send an email to users-leave@nest-simulator.org<mailto:users-leave@nest-simulator.org> -- Dr. Jochen Martin Eppler Phone: +49(2461)61-96653 ---------------------------------- Simulation Laboratory Neuroscience Jülich Supercomputing Centre Institute for Advanced Simulation ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------ Forschungszentrum Juelich GmbH 52425 Juelich Sitz der Gesellschaft: Juelich Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498 Vorsitzender des Aufsichtsrats: MinDir Volker Rieke Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender), Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt ------------------------------------------------------------------------------------------------ ------------------------------------------------------------------------------------------------