proxsuite.torch.qplayer Namespace Reference

Functions
	QPFunction (eps=1e-9, maxIter=1000, eps_backward=1.0e-4, rho_backward=1.0e-6, mu_backward=1.0e-6, omp_parallel=False, structural_feasibility=True)
	Solve a batch of Quadratic Programming (QP) problems.

Function Documentation

QPFunction()

proxsuite.torch.qplayer.QPFunction	(		eps = 1e-9,
			maxIter = 1000,
			eps_backward = 1.0e-4,
			rho_backward = 1.0e-6,
			mu_backward = 1.0e-6,
			omp_parallel = False,
			structural_feasibility = True )

Solve a batch of Quadratic Programming (QP) problems.

This function solves QP problems of the form:

$$ \begin{align} \min_{z} & ~\frac{1}{2}z^{T} Q (\theta) z +p(\theta)^{T}z \\ \text{s.t.} & ~A(\theta) z = b(\theta) \\ & ~l(\theta) \leq G(\theta) z \leq u(\theta) \end{align} $$

The QP can be infeasible - in this case the solver will return a solution to the closest feasible QP.

Parameters

[in]	eps	Tolerance for the primal infeasibility. Defaults to 1e-9.
[in]	maxIter	Maximum number of iterations. Defaults to 1000.
[in]	eps_backward	Tolerance for the backward pass. Defaults to 1e-4.
[in]	rho_backward	The new value for the primal proximal parameter. Defaults to 1e-6.
[in]	mu_backward	The new dual proximal parameter used for both equality and inequality. Defaults to 1e-6.
[in]	omp_parallel	Whether to solve the QP in parallel. Requires that proxsuite is compiled with openmp support. Defaults to False.
[in]	structural_feasibility	Whether to solve the QP with structural feasibility. Defaults to True.

Returns

QPFunctionFn or QPFunctionFn_infeas: A callable object that represents the QP problem solver. We disinguish two cases:

1. The QP is feasible. In this case, we solve the QP problem.
2. The QP is infeasible. In this case, we solve the closest feasible QP problem.

The callable object has two main methods:

Forward method

Solve the QP problem.

Parameters

[in]	Q	Batch of quadratic cost matrices of size (nBatch, n, n) or (n, n).
[in]	p	Batch of linear cost vectors of size (nBatch, n) or (n).
[in]	A	Optional batch of eq. constraint matrices of size (nBatch, p, n) or (p, n).
[in]	b	Optional batch of eq. constraint vectors of size (nBatch, p) or (p).
[in]	G	Batch of ineq. constraint matrices of size (nBatch, m, n) or (m, n).
[in]	l	Batch of ineq. lower bound vectors of size (nBatch, m) or (m).
[in]	u	Batch of ineq. upper bound vectors of size (nBatch, m) or (m).

Returns

zhats Batch of optimal primal solutions of size (nBatch, n).

lams Batch of dual variables for eq. constraint of size (nBatch, m).

nus Batch of dual variables for ineq. constraints of size (nBatch, p).

s_e Only returned in the infeasible case: batch of slack variables for eq. constraints of size (nBatch, m).

s_i Only returned in the infeasible case: batch of slack variables for ineq. constraints of size (nBatch, p).

Backward method

Compute the gradients of the QP problem with respect to its parameters.

Parameters

[in]	dl_dzhat	Batch of gradients of size (nBatch, n).
[in]	dl_dlams	Optional batch of gradients of size (nBatch, p).
[in]	dl_dnus	Optional batch of gradients of size (nBatch, m).
[in]	dl_ds_e	Only applicable in the infeasible case: optional batch of gradients of size (nBatch, m).
[in]	dl_ds_i	Only applicable in the infeasible case: optional batch of gradients of size (nBatch, m).

Returns

dQs Batch of gradients of size (nBatch, n, n).

dps Batch of gradients of size (nBatch, n).

dAs Batch of gradients of size (nBatch, p, n).

dbs Batch of gradients of size (nBatch, p).

dGs Batch of gradients of size (nBatch, m, n).

dls Batch of gradients of size (nBatch, m).

dus Batch of gradients of size (nBatch, m).

Definition at line 12 of file qplayer.py.