If I can store the cached stimulus there, then I can build and debug the Potjans D/NMNIST model version in the cloud and share the point of failure with the STDP synapses online without anyone having to install PyCall. Ie its a dependency free pathway to re-producing an error.
If I can store the cached stimulus there, then I can build and debug the Potjans D/NMNIST model version in the cloud and share the point of failure with the STDP synapses online without anyone having to install PyCall. Ie its a dependency free pathway to re-producing an error.