We have no validation files for Izhikivich neurons

[MaxStrange]

All of them are currently for LIF neurons, which is an oversight if we have the IZH model working.

[stiber]

Yes, validation is a major aspect of BrainGrid that still needs to be addressed. Test cases are the most blunt of the tools that can be used, and even the LIF test cases are arbitrary and few. A major project will be thinking in a more principled way about validation and testing, with integration into the Workbench.

From a pragmatic point of view, once we've dealt with the raiju crash issue and have done some benchmarks with the LIF model, we'll move on to IZH validation and accumulate some test cases.