Neural simulations in the Brian ecosystem

The Brian simulator, a simulator for spiking neural networks, started its life in 2007 and has been continuously developed ever since. A complete rewrite in 2014 led to the "brian2" package, which is now built as a code generation framework. This means that users write their code as a Python script, with textual representations of the model equations, and the simulator transparently translates this code into a target language such as C++. In this talk, I will give an overview over recent developments in the Brian ecosystem, showcasing a number of packages that enhance the simulator's capabilities. In particular, simulations can now be accelerated by using graphic processing units (GPUs; via the brian2genn and brian2cuda packages), and the under-development package brian2wasm brings neural simulations to the web browser. The code generation framework also makes it possible to export models as NeuroML/LEMS, or as human-readable model descriptions via the brian2tools package (which additionally offers convenience tools for visualizations). Finally, the brian2modelfitting package provides an interface to fit neuronal model parameters to experimental data, or to use simulation-based inference to estimate a parameter distribution consistent with the data. Due to Brian's architecture, this package directly benefits from improvements contributed by the other tools, such as the capability to execute simulations on a GPU.