Package TEES :: Package Detectors :: Module Detector :: Class Detector

Class Detector

Known Subclasses:

Detector is the central class of the TEES object oriented interface. Subclasses derived from it encapsulate the event and relation detection process used by TEES for the various tasks it has been developed for. When extending TEES, a new Detector can be derived from this class.

The Detector is designed for a pipeline where interaction XML is converted to machine learning examples, these examples are used to train a classifier and this classifier in turn is used to classify unknown text.

Instance Methods

[hide private]

__init__(self)

source code

__del__(self)

source code

_closeModels(self)

source code

checkStep(self, step, verbose=True)

source code

getStepStatus(self, step)

source code

setWorkDir(self, workDir='') source code

setTempWorkDir(self)

source code

deleteTempWorkDir(self)

source code

setConnection(self, connection)

source code

getModel(self)

source code

saveStr(self, name, value, model=None, modelMustExist=True)

source code

saveStrings(self, dict, model=None, modelMustExist=True)

source code

getStr(self, name, model)

source code

addClassifierModel(self, model, classifierModelPath, classifierParameters, threshold=None)

source code

openModel(self, model, mode='r') source code

getBioNLPSharedTaskParams(self, parameters=None, model=None)

source code

buildExamples(self, model, datas, outputs, golds=[], exampleStyle=None, saveIdsToModel=False, parse=None) source code

enterState(self, state, steps=None, fromStep=None, toStep=None, omitSteps=None)

source code

initVariables(self, **vars)

source code

initModel(self, model, saveParams=[]) source code

exitState(self)

source code

train(self, trainData=None, optData=None, model=None, combinedModel=None, exampleStyle=None, classifierParameters=None, parse=None, tokenization=None, fromStep=None, toStep=None)

source code

classify(self, data, model, output)

source code