#include <alg_hmc.h>
Public Member Functions | |
| AlgHmc (AlgIntAB &integrator, CommonArg &c_arg, HmcArg &arg) | |
| virtual | ~AlgHmc () |
| Performs a single HMC trajectory. | |
| Float | run (void) |
Protected Attributes | |
| int | g_size |
| < The size of the gauge field. | |
| Matrix * | gauge_field_init |
| The final gauge field configuration. | |
| Matrix * | gauge_field_final |
| The initial gauge field configuration. | |
| LRGState | lrg_state |
| The initial random numbers. | |
| int | config_no |
| Float | h_init |
| Initial Hamiltonian. | |
| Float | h_final |
| Final Hamiltonian. | |
| Float | delta_h |
| Final-Initial Hamiltonian. | |
| Float | h_delta |
| Final-Initial Hamiltonian (reversibility). | |
| Matrix * | mom |
| The (traceless antihermitian) conjugate momentum field. | |
| HmcArg * | hmc_arg |
| CommonArg * | common_arg |
| AlgIntAB * | integrator |
| The integrator which defines the algorithm. | |
| unsigned int | checksum [2] |
| Store the checksums of the final lattice (used for repro test). | |
This is an abstract implementation of HMC, the exact type of hmc is determined by the integrator that is passed to it.
The main method evolves a gauge field along a molecular dynamics trajectory followed by a metropolis accept/reject step using the integrator which is passed to it.
Definition at line 40 of file alg_hmc.h.
|
||||||||||||||||
|
< Allocate memory for the initial gauge field. < Allocate memory for the final gauge field. Definition at line 61 of file alg_hmc.C. References common_arg, cps::F_CLASS_NONE, cps::Float, cps::Verbose::Func(), cps::G_CLASS_NONE, g_size, gauge_field_final, gauge_field_init, cps::GJP, cps::Lattice::GsiteSize(), hmc_arg, integrator, cps::HmcArg::reverse, cps::smalloc(), cps::GlobalJobParameter::VolNodeSites(), and cps::VRB. |
|
|
Performs a single HMC trajectory. < Free memory for the final gauge field. Definition at line 96 of file alg_hmc.C. References cps::Verbose::Func(), gauge_field_final, gauge_field_init, hmc_arg, cps::HmcArg::reverse, cps::sfree(), and cps::VRB. |
|
|
|
Store the checksums of the final lattice (used for repro test).
Definition at line 100 of file alg_hmc.h. Referenced by run(). |
|
|
|
|
|
Definition at line 80 of file alg_hmc.h. Referenced by run(). |
|
|
Final-Initial Hamiltonian.
Definition at line 84 of file alg_hmc.h. Referenced by run(). |
|
|
< The size of the gauge field. The size of the gauge field on the local lattice in terms of the total number of floating point numbers. Definition at line 64 of file alg_hmc.h. Referenced by AlgHmc(). |
|
|
The initial gauge field configuration.
|
|
|
The final gauge field configuration.
|
|
|
Final-Initial Hamiltonian (reversibility).
Definition at line 85 of file alg_hmc.h. Referenced by run(). |
|
|
Final Hamiltonian.
Definition at line 83 of file alg_hmc.h. Referenced by run(). |
|
|
Initial Hamiltonian.
Definition at line 82 of file alg_hmc.h. Referenced by run(). |
|
|
|
|
|
The integrator which defines the algorithm.
|
|
|
The initial random numbers.
|
|
|
The (traceless antihermitian) conjugate momentum field. This is i times the actual conjugate momentum. |
1.3.9.1