Package edu.stanford.nlp.stats

A set of tools for collecting, representing, and inferring with statistics.

See: Description

Interface Summary

Interface	Description
Counter<E>	An Object to double map used for keeping weights or counts for objects.
ProbabilityDistribution<E>	This is an interface for probability measures, which will allow samples to be drawn and the probability of objects computed.
Sampler<T>	An interace for drawing samples from the label space of an object.
Scorer<L>
TwoDimensionalCounterInterface<K1,K2>	Interface representing a mapping between pairs of typed objects and double values.

Class Summary

Class	Description
AbstractCounter<E>	Default implementations of all the convenience methods provided by Counter.
AccuracyStats<L>	Utility class for aggregating counts of true positives, false positives, and false negatives and computing precision/recall/F1 stats.
ClassicCounter<E>	A specialized kind of hash table (or map) for storing numeric counts for objects.
Counters	Static methods for operating on a Counter.
Distribution<E>	Immutable class for representing normalized, smoothed discrete distributions from Counters.
GeneralizedCounter<K>	A class for keeping double counts of Lists of a prespecified length.
IntCounter<E>	A specialized kind of hash table (or map) for storing numeric counts for objects.
MultiClassAccuracyStats<L>
MultiClassChunkEvalStats	Calculates phrase based precision and recall (similar to conlleval) Handles various encodings such as IO, IOB, IOE, BILOU, SBEIO, [] Usage: java edu.stanford.nlp.stats.MultiClassChunkEvalStats [options] < filename -r - Do raw token based evaluation -d delimiter - Specifies delimiter to use (instead of tab) -b boundary - Boundary token (default is -X- ) -t defaultTag - Default tag to use if tag is not prefixed (i.e.
MultiClassPrecisionRecallExtendedStats<L>	Extension of MultiClassPrecisionRecallStats that also computes accuracy
MultiClassPrecisionRecallExtendedStats.MultiClassStringLabelStats
MultiClassPrecisionRecallExtendedStats.StringStringConverter
MultiClassPrecisionRecallStats<L>
PrecisionRecallStats	Utility class for aggregating counts of true positives, false positives, and false negatives and computing precision/recall/F1 stats.
SimpleGoodTuring	Simple Good-Turing smoothing, based on code from Sampson, available at: ftp://ftp.informatics.susx.ac.uk/pub/users/grs2/SGT.c See also http://www.grsampson.net/RGoodTur.html
TwoDimensionalCounter<K1,K2>	A class representing a mapping between pairs of typed objects and double values.
TwoDimensionalIntCounter<K1,K2>	A class representing a mapping between pairs of typed objects and int values.

Package edu.stanford.nlp.stats Description

A set of tools for collecting, representing, and inferring with statistics. The two most useful bodies of code are for (1) representing distributions (e.g., Distribution and Counter) and (2) computing the performance of a classifier (e.g., PrecisionRecallStats).

Author:

Teg Grenager