Sebastian Schmitt
Sebastian Schmitt
A Spack package is easy if: 1. the dependencies are correctly specified and are really used as dependencies and not shipped as vendor versions (git submodule or plain copies etc.)...
@mgkwill Yes, sorry for the delay.
I'm on it. Having some more questions concering the dependencies (cf. #132).
Unit tests should cover, e.g., the implementation of a FUNCTION. The test infrastructure should be such that mechanisms and tests of parts of the mechanism can be easily added also...
Please keep the tutorials and examples as pure Python files in addition. Maybe https://github.com/brian-team/brian2-binder/blob/master/generate_notebooks.py can serve as a blue print.
Cf. https://github.com/electronicvisions/brainscales2-demos where sphinx documentation is converted to jupyter notebooks.
I have a simulation where I don't exchange spikes but have many cells with different parameters. This is solved in one epoch, which makes the progress bar go from 0...
0.5 ms in simulation time? Then yes. In wall-clock time, then no.
Cf. https://brian2.readthedocs.io/en/stable/advanced/scheduling.html#progress-reporting
At the moment we don't expect to have correlated noise. We would only need an independent random number (uniform between 0 and 1) per synapse.