corpus

case class Corpus(instances: Seq[EventPair]) extends JSONSerialization with Product with Serializable

Corpus is a sequence of TrainingInstances
implicit class EventOps extends CorefMentionOps

Additional attributes and methods for a CorefMention
case class EventPair(e1: CorefMention, e2: CorefMention, relation: String, confidence: Double = AnnotationUtils.HIGH, annotatorID: String = "", notes: Option[String] = None) extends JSONSerialization with Product with Serializable

Storage class for an event pair (i.e., a single example for the relation corpus)

object AnnotationUtils

Houses values for three confidence bins, etc.
object BuildCorpus extends App with LazyLogging

Builds a relation corpus from a serialized dataset.
Builds a relation corpus from a serialized dataset.
Corpus is written as json
object BuildCorpusWithRedundancies extends App with LazyLogging

Builds a relation corpus from a serialized dataset.
Builds a relation corpus from a serialized dataset.
Corpus is written as json
object Corpus extends LazyLogging with Serializable
object CorpusBuilder extends LazyLogging

Utility functions for building the assembly relation corpus
object CorpusReader extends LazyLogging
object EventPair extends Serializable
object LegacyAnnotationReader extends LazyLogging

Attempt to recover mentions from legacy annotation format
def datasetLUT(jsonFiles: GenSeq[File]): Map[String, Vector[CorefMention]]
def datasetLUT(jsonDir: File): Map[String, Vector[CorefMention]]

Create a dataset from a directory of json files, where each file represents the reach reading results for that paper
def getPMID(docid: String): String

Retrieves PubMed ID from Document.id of an Odin Mention
def getPMID(mention: Mention): String

Retrieves PubMed ID from Document.id of an Odin Mention
lazy val logger: Logger

Attributes
protected
Definition Classes
LazyLogging