akreon-convex
Typed convex modeling, conic canonicalization, and primal-dual optimization in C++26.
Loading...
Searching...
No Matches
pd_ipm_trace.hpp File Reference
#include <array>
#include <cstddef>
#include <cstdint>
#include <type_traits>
#include "akreon/convex/solver_config.hpp"
#include "akreon/convex/canonical_trace.hpp"
#include "akreon/linalg/linalg.hpp"
Include dependency graph for pd_ipm_trace.hpp:
This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes

struct  convex::ConeStepTrace< T >
 Per-cone scalar diagnostics for one predictor or corrector pass. More...
 
struct  convex::PdIpmStepTrace< T, NumCones >
 Predictor or corrector cone diagnostics for one outer iteration. More...
 
struct  convex::PdIpmIterTrace< T, NumCones >
 Outer PD-IPM iteration diagnostics. More...
 
struct  convex::PdIpmTrace< T, NumCones >
 Fixed-capacity PD-IPM trace. More...
 
struct  convex::EmptyPdIpmSnapshot
 Empty object used when full snapshot tracing is disabled. More...
 
struct  convex::PdIpmConeSnapshot< T, MaxConeDim >
 Cone snap shot at current iterate. More...
 
struct  convex::PdIpmGlobalStepSnapshot< T, VarDim, EqDim >
 Global snap shot at current iterate. More...
 
struct  convex::PdIpmFullIterSnapshot< T, NumCones, MaxConeDim, VarDim, EqDim >
 Optional full debug snapshot for one PD-IPM iteration. More...
 

Namespaces

namespace  convex
 

Typedefs

template<bool Enable, linalg::Field T, std::size_t NumCones, std::size_t MaxConeDim, std::size_t VarDim, std::size_t EqDim>
using convex::PdIpmSnapshotStorage = std::conditional_t< Enable, PdIpmFullIterSnapshot< T, NumCones, MaxConeDim, VarDim, EqDim >, EmptyPdIpmSnapshot >
 Alias for snap shot storage (full or empty)
 
template<linalg::Field T, typename CanonicalTraceT >
using convex::PdIpmTraceForCanonicalTrace = PdIpmTrace< T, canonical_trace_num_cones_v< CanonicalTraceT > >
 Convenience alias for a solver trace matched to a CanonicalTrace.