protected class SimpleDistribution.Trainer extends Object implements DistributionTrainer
Constructor and Description |
---|
SimpleDistribution.Trainer()
Create a new trainer.
|
Modifier and Type | Method and Description |
---|---|
void |
addCount(DistributionTrainerContext dtc,
AtomicSymbol sym,
double times)
Registers that sym was counted in this state.
|
void |
clearCounts(DistributionTrainerContext dtc)
Clears all of the counts to zero.
|
double |
getCount(DistributionTrainerContext dtc,
AtomicSymbol sym)
Get the current count for this state.
|
void |
train(DistributionTrainerContext dtc,
double weight)
Trains the Distribution, given a null model.
|
protected void |
trainImpl(DistributionTrainerContext dtc,
double weight) |
public SimpleDistribution.Trainer()
public void addCount(DistributionTrainerContext dtc, AtomicSymbol sym, double times) throws IllegalSymbolException
DistributionTrainer
Registers that sym was counted in this state.
This method may be called multiple times with the same symbol. In this case, the times should be summed.
addCount
in interface DistributionTrainer
dtc
- the DistributionTrainerContext within which the count was addedsym
- the Symbol seentimes
- the number of times to addIllegalSymbolException
- if sym is not recognisedpublic double getCount(DistributionTrainerContext dtc, AtomicSymbol sym) throws IllegalSymbolException
DistributionTrainer
Get the current count for this state.
This method may be called multiple times with the same symbol. Each time it should return the agregate of the counts added with addCount since the last invocation of clearCounts.
getCount
in interface DistributionTrainer
dtc
- the DistributionTrainerContext within which the count was addedsym
- the Symbol seenIllegalSymbolException
- if sym is not recognisedpublic void clearCounts(DistributionTrainerContext dtc)
DistributionTrainer
clearCounts
in interface DistributionTrainer
public void train(DistributionTrainerContext dtc, double weight) throws ChangeVetoException
DistributionTrainer
Trains the Distribution, given a null model.
This will use the information collected with multiple addCount calls, and the null model to generate the new weights.
This method should not modify the underlying counts.
train
in interface DistributionTrainer
dtc
- the context to useweight
- how many lots of the null model to addChangeVetoException
- if the distribution could not have its weights
modifiedprotected void trainImpl(DistributionTrainerContext dtc, double weight)
Copyright © 2014 BioJava. All rights reserved.