edu.stanford.nlp.tagger.maxent
Class LambdaSolveTagger
java.lang.Object
|
+--edu.stanford.nlp.maxent.iis.LambdaSolve
|
+--edu.stanford.nlp.tagger.maxent.LambdaSolveTagger
- public class LambdaSolveTagger
- extends LambdaSolve
This module does the working out of lambda parameters for binary tagger
features. It can use either IIS or CG.
Fields inherited from class edu.stanford.nlp.maxent.iis.LambdaSolve |
eps, fixedFnumXY, fnumArr, ftieldArr, lambda, lambda_converged, newtonerr, p, probConds, smooth, zlambda |
Method Summary |
boolean |
checkCorrectness()
Check whether the constraints are satisfied, the probabilities sum to one, etc. |
double[] |
getDerivatives()
assuming we have the lambdas in the array and we need only the
derivatives now. |
double |
logLikelihoodScratch()
calculate the log likelihood from scratch, hashing the conditional
probabilities in pcond
which we will use for the derivative later. |
void |
readOldLambdas(String filename,
String oldfilename)
|
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
readOldLambdas
public void readOldLambdas(String filename,
String oldfilename)
checkCorrectness
public boolean checkCorrectness()
- Description copied from class:
LambdaSolve
- Check whether the constraints are satisfied, the probabilities sum to one, etc. Prints out a message
if there is something wrong.
- Overrides:
checkCorrectness
in class LambdaSolve
logLikelihoodScratch
public double logLikelihoodScratch()
- calculate the log likelihood from scratch, hashing the conditional
probabilities in pcond
which we will use for the derivative later.
- Overrides:
logLikelihoodScratch
in class LambdaSolve
getDerivatives
public double[] getDerivatives()
- assuming we have the lambdas in the array and we need only the
derivatives now.
- Overrides:
getDerivatives
in class LambdaSolve
Stanford NLP Group