org.opencv.ml
public class CvANN_MLP_TrainParams extends java.lang.Object
Parameters of the MLP training algorithm. You can initialize the structure by a constructor or the individual parameters can be adjusted after the structure is created.
The back-propagation algorithm parameters:
Strength of the weight gradient term. The recommended value is about 0.1.
Strength of the momentum term (the difference between weights on the 2
previous iterations). This parameter provides some inertia to smooth the
random fluctuations of the weights. It can vary from 0 (the feature is
disabled) to 1 and beyond. The value 0.1 or so is good enough
// C++ code:
The RPROP algorithm parameters (see [RPROP93] for details):
Initial value Delta_0 of update-values Delta_(ij).
Increase factor eta^+. It must be >1.
Decrease factor eta^-. It must be <1.
Update-values lower limit Delta_(min). It must be positive.
Update-values upper limit Delta_(max). It must be >1.
Constructor and Description |
---|
CvANN_MLP_TrainParams()
The constructors.
|
Modifier and Type | Method and Description |
---|---|
double |
get_bp_dw_scale() |
double |
get_bp_moment_scale() |
double |
get_rp_dw_max() |
double |
get_rp_dw_min() |
double |
get_rp_dw_minus() |
double |
get_rp_dw_plus() |
double |
get_rp_dw0() |
TermCriteria |
get_term_crit() |
int |
get_train_method() |
void |
set_bp_dw_scale(double bp_dw_scale) |
void |
set_bp_moment_scale(double bp_moment_scale) |
void |
set_rp_dw_max(double rp_dw_max) |
void |
set_rp_dw_min(double rp_dw_min) |
void |
set_rp_dw_minus(double rp_dw_minus) |
void |
set_rp_dw_plus(double rp_dw_plus) |
void |
set_rp_dw0(double rp_dw0) |
void |
set_term_crit(TermCriteria term_crit) |
void |
set_train_method(int train_method) |
public static final int BACKPROP
public static final int RPROP
public CvANN_MLP_TrainParams()
The constructors.
By default the RPROP algorithm is used:
// C++ code:
CvANN_MLP_TrainParams.CvANN_MLP_TrainParams()
term_crit = cvTermCriteria(CV_TERMCRIT_ITER + CV_TERMCRIT_EPS, 1000, 0.01);
train_method = RPROP;
bp_dw_scale = bp_moment_scale = 0.1;
rp_dw0 = 0.1; rp_dw_plus = 1.2; rp_dw_minus = 0.5;
rp_dw_min = FLT_EPSILON; rp_dw_max = 50.;
public double get_bp_dw_scale()
public double get_bp_moment_scale()
public double get_rp_dw_max()
public double get_rp_dw_min()
public double get_rp_dw_minus()
public double get_rp_dw_plus()
public double get_rp_dw0()
public TermCriteria get_term_crit()
public int get_train_method()
public void set_bp_dw_scale(double bp_dw_scale)
public void set_bp_moment_scale(double bp_moment_scale)
public void set_rp_dw_max(double rp_dw_max)
public void set_rp_dw_min(double rp_dw_min)
public void set_rp_dw_minus(double rp_dw_minus)
public void set_rp_dw_plus(double rp_dw_plus)
public void set_rp_dw0(double rp_dw0)
public void set_term_crit(TermCriteria term_crit)
public void set_train_method(int train_method)