TEESTrain

Synopsis

Train a model that can be used to predict binary relations using TEES

Description

Snippet

📋

<teestrain class="TEESTrain">
    <modelTargetDir></modelTargetDir>
    <namedEntityLayer></namedEntityLayer>
    <python2Executable></python2Executable>
    <schema></schema>
    <teesHome></teesHome>
</teestrain>

Mandatory parameters

modelTargetDir

Path to the directory where put the trained model

namedEntityLayer

Name of the layer containing the named entities.

python2Executable

Path to the Python 2 executable, TEES will fail if run through Python 3.

schema

Give the schema of the relations to train i.e.

      	  <schema>
	    <Lives_In>Bacteria,Location</Lives_In>
      	  </schema>
	

teesHome

Path to the TEES home directory.

Optional parameters

constantRelationFeatures

Constant features to add to each relation created by this module.

constantTupleFeatures

Constant features to add to each tuple created by this module.

corpusSetFeature

UNDOCUMENTED

devSetValue

Feature key of the dev set corpus.

documentFilter

Only process document that satisfy this expression.

modelName

give a name to the trained model

namedEntityTypeFeature

Name of the feature to access the type of the named entities.

omitSteps

Set the preprocessing steps to ignore in the form of [GENIA_SPLITTER][,BANNER][,BLLIP_BIO][,STANFORD_CONVERT][,SPLIT_NAMES][,FIND_HEADS]

posFeature

Name of the feature containing the POS-tag.

sectionFilter

Process only sections that satisfy this expression.

sentenceLayer

Name of the layer that contains sentence annotations.

testSetValue

Feature key of the test set corpus.

tokenLayer

Name of the layer that contains tokens.

trainSetValue

Feature key of the train set corpus.

Deprecated parameters

namedEntityLayerName

Deprecated alias for namedEntityLayer .

sentenceLayerName

Deprecated alias for sentenceLayer .

tokenLayerName

Deprecated alias for tokenLayer .