hex
Class DGLM.GLMValidation
java.lang.Object
water.Iced
hex.DGLM.GLMValidation
- All Implemented Interfaces:
- java.lang.Cloneable, Freezable
- Enclosing class:
- DGLM
public static class DGLM.GLMValidation
- extends Iced
Methods inherited from class java.lang.Object |
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
_modelKeys
public final Key[] _modelKeys
KEY_PREFIX
public static final java.lang.String KEY_PREFIX
- See Also:
- Constant Field Values
_n
public long _n
_xvalIterations
public int _xvalIterations
_caseCount
public long _caseCount
_aic
public double _aic
_deviance
public double _deviance
_nullDeviance
public double _nullDeviance
_err
public double _err
_cm
public ConfusionMatrix[] _cm
DGLM.GLMValidation
public DGLM.GLMValidation()
DGLM.GLMValidation
public DGLM.GLMValidation(Key modelKey,
Key[] modelKeys,
ConfusionMatrix.ErrMetric m,
double[] thresholds,
long time)
computationTime
public final long computationTime()
dataKey
public Key dataKey()
modelKey
public Key modelKey()
models
public java.lang.Iterable<DGLM.GLMModel> models()
fold
public int fold()
bestCM
public ConfusionMatrix bestCM()
err
public double err()
bestCM
public ConfusionMatrix bestCM(ConfusionMatrix.ErrMetric errM)
bestThreshold
public double bestThreshold()
computeBestThreshold
public void computeBestThreshold(ConfusionMatrix.ErrMetric errM)
classError
public double[] classError()
AUC
public double AUC()
computeAUC
protected void computeAUC()
- Computes area under the ROC curve. The ROC curve is computed from the confusion matrices
(there is one for each computed threshold). Area under this curve is then computed as a sum
of areas of trapezoids formed by each neighboring points.
toJson
public com.google.gson.JsonObject toJson()
AIC
public double AIC()