BrainPy is a highly flexible and extensible framework targeting on the high-performance Brain Dynamics Programming (BDP). Among its key ingredients, BrainPy supports:
JIT compilation and automatic differentiation for class objects.
Numerical methods for ordinary differential equations (ODEs), stochastic differential equations (SDEs), delay differential equations (DDEs), fractional differential equations (FDEs), etc.
Dynamics building with the modular and composable programming interface.
Dynamics simulation for various brain objects with parallel supports.
Dynamics training with various machine learning algorithms, like FORCE learning, ridge regression, back-propagation, etc.
Dynamics analysis for low- and high-dimensional systems, including phase plane analysis, bifurcation analysis, linearization analysis, and fixed/slow point finding.
And more others ……
Comprehensive examples of BrainPy please see:
The code of BrainPy is open-sourced at GitHub:
BrainPy version 2.2.x has been released. See release note for details about changes from brainpy 2.1.x.
- Math Basics
- Model Building
- Model Simulation
- Model Training
- Model Analysis
- Numerical Solvers for Ordinary Differential Equations
- Numerical Solvers for Stochastic Differential Equations
- Numerical Solvers for Fractional Differential Equations
- Numerical Solvers for Delay Differential Equations
- Joint Differential Equations
- Synaptic Connections
- Synaptic Weights
- Gradient Descent Optimizers
- Saving and Loading
- Inputs Construction
- Release notes (brainpy)