📚 API Reference

This reference provides a detailed overview of the functions and classes in the Archimedes library.

Modules

observers

State estimation and Kalman filtering

optimize

Numerical optimization algorithms

sysid

System identification and parameter estimation functionality

tree

Utilities for working with hierarchical "pytree" data structures.

tree.struct

Utilities for defining custom classes that can be used with pytree transformations.

Core Functions

array(x[, dtype])

Create an array supporting both numeric and symbolic computation.

codegen(func, args[, static_argnums, ...])

Generate C/C++ code from a compiled function.

discretize([func, dt, method, n_steps, name])

Convert continuous-time dynamics to discrete-time using numerical integration.

sym(name[, shape, dtype, kind])

Create a symbolic array for use in symbolic computations.

sym_like(x, name[, dtype, kind])

Create a symbolic array with the same shape and dtype as an existing array.

zeros(shape[, dtype, sparse, kind])

Construct a symbolic array of zeros with the given shape and dtype.

ones(shape[, dtype, kind])

Construct a symbolic array of ones with the given shape and dtype.

zeros_like(x[, dtype, sparse, kind])

Create a symbolic array of zeros with the same shape and dtype as an input array.

ones_like(x[, dtype, kind])

Create a symbolic array of ones with the same shape and dtype as an input array.

eye(N[, M, k, dtype, order, device, kind])

Construct a symbolic identity matrix of size n with the given dtype.

scan(func, init_carry[, xs, length])

Apply a function repeatedly while carrying state between iterations.

switch(index, branches, *args[, name, kind])

Selectively apply one of several functions based on an index.

vmap(func[, in_axes, out_axes, name])

Vectorize a function along specified argument axes.

compile([func, static_argnums, ...])

Create a "compiled" function from a Python function.

callback(func, result_shape_dtypes, *args)

Execute an arbitrary Python function within an symbolic computational graph.

grad(func[, argnums, name, static_argnums, ...])

Create a function that evaluates the gradient of func.

jac(func[, argnums, name, static_argnums, ...])

Create a function that evaluates the Jacobian of func.

hess(func[, argnums, name, static_argnums, ...])

Create a function that evaluates the Hessian of func.

jvp(func[, name, static_argnums, ...])

Create a function that evaluates the Jacobian-vector product of func.

vjp(func[, name, static_argnums, ...])

Create a function that evaluates the vector-Jacobian product of func.

interpolant(grid, data[, method, arg_names, ...])

Create a callable N-dimensional interpolant function.

integrator(func[, method, atol, rtol, ...])

Create an ODE solver function from a dynamics function.

odeint(func, t_span, x0[, method, t_eval, ...])

Integrate a system of ordinary differential equations.

nlp_solver(obj[, constr, static_argnames, ...])

Create a reusable solver for a nonlinear optimization problem.

minimize(obj, x0[, args, static_argnames, ...])

Minimize a scalar function with optional constraints and PyTree support.

implicit(func[, static_argnames, solver, name])

Construct an explicit function from an implicit relation.

root(func, x0[, args, static_argnames, ...])

Find a root of a nonlinear function.

qpsol(obj, constr, x0[, lba, uba, lam_a0, ...])

Solve a quadratic programming problem