alpaqa 0.0.1
Nonconvex constrained optimization
Static Public Member Functions | List of all members
PANOCDirection< DirectionProviderT > Struct Template Reference

#include <alpaqa/inner/directions/decl/panoc-direction-update.hpp>

Detailed Description

template<class DirectionProviderT>
struct alpaqa::PANOCDirection< DirectionProviderT >

Definition at line 8 of file panoc-direction-update.hpp.

+ Collaboration diagram for PANOCDirection< DirectionProviderT >:

Static Public Member Functions

static void initialize (DirectionProviderT &dp, crvec x₀, crvec x̂₀, crvec p₀, crvec grad₀)=delete
 
static bool update (DirectionProviderT &dp, crvec xₖ, crvec xₖ₊₁, crvec pₖ, crvec pₖ₊₁, crvec gradₖ₊₁, const Box &C, real_t γₖ₊₁)=delete
 
static bool apply (DirectionProviderT &dp, crvec xₖ, crvec x̂ₖ, crvec pₖ, real_t γ, rvec qₖ)=delete
 Apply the direction estimation in the current point. More...
 
static void changed_γ (DirectionProviderT &dp, real_t γₖ, real_t old_γₖ)=delete
 

Member Function Documentation

◆ initialize()

static void initialize ( DirectionProviderT &  dp,
crvec  x₀,
crvec  x̂₀,
crvec  p₀,
crvec  grad₀ 
)
staticdelete

◆ update()

static bool update ( DirectionProviderT &  dp,
crvec  xₖ,
crvec  xₖ₊₁,
crvec  pₖ,
crvec  pₖ₊₁,
crvec  gradₖ₊₁,
const Box C,
real_t  γₖ₊₁ 
)
staticdelete

◆ apply()

static bool apply ( DirectionProviderT &  dp,
crvec  xₖ,
crvec  x̂ₖ,
crvec  pₖ,
real_t  γ,
rvec  qₖ 
)
staticdelete

Apply the direction estimation in the current point.

Parameters
[in]dpDirection provider (e.g. LBFGS, Anderson Acceleration).
[in]xₖCurrent iterate.
[in]x̂ₖResult of proximal gradient step in current iterate.
[in]pₖProximal gradient step between x̂ₖ and xₖ.
[in]γH₀ = γI for L-BFGS
[out]qₖResulting step.

◆ changed_γ()

static void changed_γ ( DirectionProviderT &  dp,
real_t  γₖ,
real_t  old_γₖ 
)
staticdelete

The documentation for this struct was generated from the following file: