Cleaning up help text for the -numBad argument.
This commit is contained in:
parent
163d796ae7
commit
59f56bef30
|
|
@ -264,7 +264,7 @@ public class VariantDataManager {
|
||||||
Collections.sort( data, new VariantDatum.VariantDatumLODComparator() );
|
Collections.sort( data, new VariantDatum.VariantDatumLODComparator() );
|
||||||
final int numToAdd = minimumNumber - trainingData.size();
|
final int numToAdd = minimumNumber - trainingData.size();
|
||||||
if( numToAdd > data.size() ) {
|
if( numToAdd > data.size() ) {
|
||||||
throw new UserException.BadInput( "Error during negative model training. Minimum number of variants to use in training is larger than the whole call set. One can attempt to lower the --minNumBadVariants arugment but this is unsafe." );
|
throw new UserException.BadInput( "Error during negative model training. Minimum number of variants to use in training is larger than the whole call set. One can attempt to lower the --numBadVariants arugment but this is unsafe." );
|
||||||
}
|
}
|
||||||
int index = 0, numAdded = 0;
|
int index = 0, numAdded = 0;
|
||||||
while( numAdded < numToAdd && index < data.size() ) {
|
while( numAdded < numToAdd && index < data.size() ) {
|
||||||
|
|
@ -275,7 +275,7 @@ public class VariantDataManager {
|
||||||
numAdded++;
|
numAdded++;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
logger.info( "Additionally training with worst " + numToAdd + "% of passing data --> " + (trainingData.size() - numBadSitesAdded) + " variants with LOD <= " + String.format("%.4f", data.get(index).lod) + "." );
|
logger.info( "Additionally training with worst " + numToAdd + " scoring variants --> " + (trainingData.size() - numBadSitesAdded) + " variants with LOD <= " + String.format("%.4f", data.get(index).lod) + "." );
|
||||||
return trainingData;
|
return trainingData;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -335,7 +335,7 @@ public class VariantRecalibrator extends RodWalker<ExpandingArrayList<VariantDat
|
||||||
engine.evaluateData( dataManager.getData(), badModel, true );
|
engine.evaluateData( dataManager.getData(), badModel, true );
|
||||||
|
|
||||||
if( badModel.failedToConverge || goodModel.failedToConverge ) {
|
if( badModel.failedToConverge || goodModel.failedToConverge ) {
|
||||||
throw new UserException("NaN LOD value assigned. Clustering with this few variants and these annotations is unsafe. Please consider raising the number of variants used to train the negative model (via --minNumBad, for example) or lowering the maximum number of Gaussians to use in the model (via --maxGaussians 4, for example)");
|
throw new UserException("NaN LOD value assigned. Clustering with this few variants and these annotations is unsafe. Please consider " + (badModel.failedToConverge ? "raising the number of variants used to train the negative model (via --numBad 3000, for example)." : "lowering the maximum number of Gaussians allowed for use in the model (via --maxGaussians 4, for example).") );
|
||||||
}
|
}
|
||||||
|
|
||||||
engine.calculateWorstPerformingAnnotation( dataManager.getData(), goodModel, badModel );
|
engine.calculateWorstPerformingAnnotation( dataManager.getData(), goodModel, badModel );
|
||||||
|
|
|
||||||
|
|
@ -94,6 +94,6 @@ public class VariantRecalibratorArgumentCollection {
|
||||||
public double DIRICHLET_PARAMETER = 0.001;
|
public double DIRICHLET_PARAMETER = 0.001;
|
||||||
@Argument(fullName="priorCounts", shortName="priorCounts", doc="The number of prior counts to use in the variational Bayes algorithm.", required=false)
|
@Argument(fullName="priorCounts", shortName="priorCounts", doc="The number of prior counts to use in the variational Bayes algorithm.", required=false)
|
||||||
public double PRIOR_COUNTS = 20.0;
|
public double PRIOR_COUNTS = 20.0;
|
||||||
@Argument(fullName="numBadVariants", shortName="numBad", doc="The number of worst scoring variants to use when building the Gaussian mixture model of bad variants. Will override -percentBad argument if necessary.", required=false)
|
@Argument(fullName="numBadVariants", shortName="numBad", doc="The number of worst scoring variants to use when building the Gaussian mixture model of bad variants.", required=false)
|
||||||
public int NUM_BAD_VARIANTS = 1000;
|
public int NUM_BAD_VARIANTS = 1000;
|
||||||
}
|
}
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue