ALGORITHM The main implementation of the LRFG algorithm This algorithm is an implementation of the low-rank factor greedy algorithm as it was presented in the paper by A. Schmidt and B. Haasdonk (2016). It is a generic class which can be applied to all OOP models that implement a certain interface.
Definition at line 19 of file Algorithm.m.
Public Member Functions | |
Algorithm (model,ModelData model_data, M_train) | |
ALGORITHM Constructor for the LRFG algorithm. More... | |
function [
V , W , info ] = | run () |
================================================================= RUN Implementation of the LRFG algorithm. This function actually performs the algorithm and fills the corresponding fields in detailed_data More... | |
function [
errs_normalized , errs_absolute , avg_time ] = | calc_error_indicator (RB_W, RB_V, mus) |
Calculate the errors for the basis stored in the matrix RB. More... | |
Public Attributes | |
M_train | |
Save the M_train class for subsequent use. More... | |
error_indicator = "normalized_residual" | |
Use the normalized residual as error indicator. More... | |
orthonormalize_E = true | |
Orthonormalize the basis wrt. to the mass matrix. More... | |
Protected Member Functions | |
function this = | initialize_basis (model,ModelData model_data) |
This function should initialize the basis with a first element. The default implementation takes the solution. More... | |
Protected Attributes | |
model | |
Pointer to the ReducedModel class. | |
model_data | |
The reference to the ModelData class. | |
Greedy.LRFG.Algorithm.Algorithm | ( | model, | |
ModelData | model_data, | ||
M_train | |||
) |
ALGORITHM Constructor for the LRFG algorithm.
model | model |
model_data | Matlab structure storing (possibly) high dimensional data needed by IDetailedModel.detailed_simulation(). |
M_train | M train |
Definition at line 74 of file Algorithm.m.
function [ errs_normalized , errs_absolute , avg_time ] = Greedy.LRFG.Algorithm.calc_error_indicator | ( | RB_W, | |
RB_V, | |||
mus | |||
) |
Calculate the errors for the basis stored in the matrix RB.
RB_W | RB W |
RB_V | RB V |
mus | mus |
errs_normalized | errs normalized |
errs_absolute | errs absolute |
avg_time | avg time |
Definition at line 355 of file Algorithm.m.
|
protected |
This function should initialize the basis with a first element. The default implementation takes the solution.
model | model |
model_data | Matlab structure storing (possibly) high dimensional data needed by IDetailedModel.detailed_simulation(). |
this | this |
set_mu —
set muRB_V —
RB V RB_W —
RB W Definition at line 113 of file Algorithm.m.
function [ V , W , info ] = Greedy.LRFG.Algorithm.run | ( | ) |
================================================================= RUN Implementation of the LRFG algorithm. This function actually performs the algorithm and fills the corresponding fields in detailed_data
V | V |
W | W |
info | info |
greedy_tolerance —
greedy tolerance orthonormalize_E —
orthonormalize E error_indicator —
error indicator pod_max_extension —
pod max extension pod_tolerance —
pod tolerance Z_sizes —
Z sizes vectors_added —
vectors added detailed_simulation_times —
detailed simulation times detailed_simulation_times_real —
detailed simulation times real chosen_mu —
chosen mu error_history_normalized —
error history normalized error_history_absolute —
error history absolute error_decay —
error decay num_solves —
num solves M_train —
M train runtime —
runtime Definition at line 154 of file Algorithm.m.
Greedy.LRFG.Algorithm.error_indicator = "normalized_residual" |
Use the normalized residual as error indicator.
Default: "normalized_residual"
Definition at line 55 of file Algorithm.m.
Greedy.LRFG.Algorithm.M_train |
Save the M_train class for subsequent use.
Definition at line 48 of file Algorithm.m.
Greedy.LRFG.Algorithm.orthonormalize_E = true |
Orthonormalize the basis wrt. to the mass matrix.
Default: true
Definition at line 64 of file Algorithm.m.