pymc.ode.DifferentialEquation#

class pymc.ode.DifferentialEquation(func, times, *, n_states, n_theta, t0=0)[source]#

Specify an ordinary differential equation

Due to the nature of the model (as well as included solvers), the process of ODE solution may perform slowly. A faster alternative library based on PyMC–sunode–has implemented Adams’ method and BDF (backward differentation formula). More information about sunode is available at: aseyboldt/sunode.

\[\dfrac{dy}{dt} = f(y,t,p) \quad y(t_0) = y_0\]

Parameters:

funccallable(): Function specifying the differential equation. Must take arguments y (n_states,), t (scalar), p (n_theta,)
timesarray: Array of times at which to evaluate the solution of the differential equation.
n_statesint: Dimension of the differential equation. For scalar differential equations, n_states=1. For vector valued differential equations, n_states = number of differential equations in the system.
n_thetaint: Number of parameters in the differential equation.
t0float: Time corresponding to the initial condition

Examples

def odefunc(y, t, p):
    # Logistic differential equation
    return p[0] * y[0] * (1 - y[0])

times = np.arange(0.5, 5, 0.5)

ode_model = DifferentialEquation(func=odefunc, times=times, n_states=1, n_theta=1, t0=0)

Methods

`DifferentialEquation.L_op`(inputs, outputs, ...)	Construct a graph for the L-operator.
`DifferentialEquation.R_op`(inputs, eval_points)	Construct a graph for the R-operator.
`DifferentialEquation.__init__`(func, times, ...)
`DifferentialEquation.add_tag_trace`(thing[, ...])	Add tag.trace to a node or variable.
`DifferentialEquation.do_constant_folding`(...)	Determine whether or not constant folding should be performed for the given node.
`DifferentialEquation.grad`(inputs, output_grads)	Construct a graph for the gradient with respect to each input variable.
`DifferentialEquation.infer_shape`(fgraph, ...)
`DifferentialEquation.make_node`(y0, theta)	Construct an Apply node that represent the application of this operation to the given inputs.
`DifferentialEquation.make_py_thunk`(node, ...)	Make a Python thunk.
`DifferentialEquation.make_thunk`(node, ...[, ...])	Create a thunk.
`DifferentialEquation.perform`(node, ...)	Calculate the function on the inputs and put the variables in the output storage.
`DifferentialEquation.prepare_node`(node, ...)	Make any special modifications that the Op needs before doing `Op.make_thunk()`.

Attributes

`default_output`	An `int` that specifies which output `Op.__call__()` should return.
`destroy_map`	A `dict` that maps output indices to the input indices upon which they operate in-place.
`itypes`
`otypes`
`view_map`	A `dict` that maps output indices to the input indices of which they are a view.