brainpy module#

Numerical Differential Integration#

Integrator(variables, parameters, arguments, dt)

Basic Integrator Class.


Make a joint equation from multiple derivation functions.

IntegratorRunner(target[, inits, dt, ...])

Structural runner for numerical integrators in brainpy.

odeint([f, method, var_type, dt, name, ...])

Numerical integration for ODEs.

sdeint([f, g, method, dt, name, show_code, ...])

Numerical integration for SDEs.

fdeint(alpha, num_memory, inits[, f, ...])

Numerical integration for FDEs.

Building Dynamical System#

DynamicalSystem([name, mode])

Base Dynamical System class.

Container(*dynamical_systems_as_tuple[, ...])

Container object which is designed to add other instances of DynamicalSystem.

Sequential(*modules_as_tuple[, name, mode])

A sequential input-output module.

Network(*ds_tuple[, name, mode])

Base class to model network objects, an alias of Container.

NeuGroup(size[, keep_size, name, mode])

Base class to model neuronal groups.

SynConn(pre, post[, conn, name, mode])

Base class to model two-end synaptic connections.

SynOut([name, target_var])

Base class for synaptic current output.

SynSTP(*args, **kwargs)

Base class for synaptic short-term plasticity.

SynLTP(*args, **kwargs)

Base class for synaptic long-term plasticity.

TwoEndConn(pre, post[, conn, output, stp, ...])

Base class to model synaptic connections.

CondNeuGroup(size[, keep_size, C, A, V_th, ...])

Base class to model conductance-based neuron group.

Channel(size[, name, keep_size, mode])

Abstract channel class.

Simulating Dynamical System#

DSRunner(target[, inputs, monitors, ...])

The runner for DynamicalSystem.

Training Dynamical System#

DSTrainer(target, **kwargs)

Structural Trainer for Dynamical Systems.

BPTT(target, loss_fun[, optimizer, ...])

The trainer implementing the back-propagation through time (BPTT) algorithm for training dyamical systems.

BPFF(target, loss_fun[, optimizer, ...])

The trainer implementing back propagation algorithm for feedforward neural networks.

OnlineTrainer(target[, fit_method])

Online trainer for models with recurrent dynamics.

ForceTrainer(target[, alpha])

FORCE learning.

OfflineTrainer(target[, fit_method])

Offline trainer for models with recurrent dynamics.

RidgeTrainer(target[, alpha])

Trainer of ridge regression, also known as regression with Tikhonov regularization.

Dynamical System Helpers#

LoopOverTime(target[, out_vars, no_state, ...])

Transform a single step DynamicalSystem into a multiple-step forward propagation BrainPyObject.