brian2cuda
brian2cuda copied to clipboard
A brian2 extension to simulate spiking neural networks on GPUs
For the first release, we won't support Windows. For Windows support, I need to figure out how CUDA is installed on Windows and how to write Windows makefiles. And find...
This check is only required when one of the `SynapticPathway` objects of the `Synapses` object does not use atomics and has synaptic effect mode `target`. Since we use atomics by...
This PR 1. Removes most of the INFO prints during `cuda_standalone` simulations. They are now debug prints. 2. Adds a simple logging system for the generated code, with the same...
I just realized that the Brian2 update (#249 ) broke the benchmark scripts because some changes in `brian2.diff` got lost. See my comment here: https://github.com/brian-team/brian2cuda/pull/249#discussion_r1008176366
`SpikeGeneratorClock(..., dt=2*defaultclock.dt)` produces wrong results in the following test: https://github.com/brian-team/brian2cuda/blob/2b28ac6606544b64185b671185edd9dd7332667a/brian2cuda/tests/test_synaptic_propagations.py#L231-L266 This could be a general issue of how we compute the circular eventspaces and might also apply to a normal...
Currently our `EventMonitor` (~ `SpikeMonitor`) returns times and indices sorted by time but for each time step the indices are not sorted. So we get e.g. this ```python times =...
This is a follow-up of #83 In PR #300, I fixed the spike loss between `run` calls in general (when there are delays and spikes are still in queues or...
Implement [this brian2 test](https://github.com/brian-team/brian2/blob/320a5f6b36e2aa5dfe7e6b22bd46c3652ad9cc99/brian2/tests/test_synapses.py#L1084-L1106) for brian2cuda. A first test did not show any spikes in cuda, but some in cpp (but also not all that are tested for). I just...
There seems to be a memory leak when having multiple `run` calls, or at least the memory used on the GPU constantly increases. - In PR #300, I removed some...