brainpy.math.autograd.hessian#

brainpy.math.autograd.hessian(func, dyn_vars=None, grad_vars=None, argnums=None, holomorphic=False, return_value=False)[source]#

Hessian of func as a dense array.

Parameters

  • func (callable, function) – Function whose Hessian is to be computed. Its arguments at positions specified by argnums should be arrays, scalars, or standard Python containers thereof. It should return arrays, scalars, or standard Python containers thereof.

  • dyn_vars (optional, ArrayCollector, sequence of JaxArray) – The dynamical changed variables.

  • grad_vars (optional, ArrayCollector, sequence of JaxArray) – The variables required to compute their gradients.

  • argnums (Optional, integer or sequence of integers) – Specifies which positional argument(s) to differentiate with respect to (default 0).

  • holomorphic (bool) – Indicates whether fun is promised to be holomorphic. Default False.

  • return_value (bool) – Whether return the hessian values.