edu.emory.mathcs.nlp.configuration.config-train-pos.xml Maven / Gradle / Ivy
The newest version!
<!-- part-of-speech tagging --> <configuration> <tsv> <column index="1" field="form"/> <column index="3" field="pos"/> </tsv> <lexica> <ambiguity_classes field="word_form_simplified_lowercase">edu/emory/mathcs/nlp/lexica/en-ambiguity-classes-simplified-lowercase.xz</ambiguity_classes> <word_clusters field="word_form_simplified_lowercase">edu/emory/mathcs/nlp/lexica/en-brown-clusters-simplified-lowercase.xz</word_clusters> </lexica> <optimizer> <l1_regularization>0.00001</l1_regularization> <algorithm>adagrad-mini-batch</algorithm> <learning_rate>0.02</learning_rate> <feature_cutoff>2</feature_cutoff> <lols fixed="0" decaying="0.95"/> <batch_size>5</batch_size> <max_epoch>40</max_epoch> <bias>0</bias> </optimizer> <reducer> <lower_bound>97.48</lower_bound> <increment>0.01</increment> <iteration>2</iteration> <start>0.05</start> <range>0.005</range> </reducer> <feature_template> <!-- 1-gram features --> <feature f0="i-2:word_form_simplified"/> <feature f0="i-1:word_form_simplified"/> <feature f0="i:word_form_simplified"/> <feature f0="i+1:word_form_simplified"/> <feature f0="i+2:word_form_simplified"/> <feature f0="i-1:word_shape"/> <feature f0="i:word_shape"/> <feature f0="i+1:word_shape"/> <feature f0="i-3:part_of_speech_tag"/> <feature f0="i-2:part_of_speech_tag"/> <feature f0="i-1:part_of_speech_tag"/> <feature set="true" f0="i:ambiguity_classes"/> <feature set="true" f0="i+1:ambiguity_classes"/> <feature set="true" f0="i+2:ambiguity_classes"/> <feature set="true" f0="i+3:ambiguity_classes"/> <!-- 2-gram features --> <feature f0="i-2:word_form_simplified_lowercase" f1="i-1:word_form_simplified_lowercase"/> <feature f0="i-1:word_form_simplified_lowercase" f1="i:word_form_simplified_lowercase"/> <feature f0="i:word_form_simplified_lowercase" f1="i+1:word_form_simplified_lowercase"/> <feature f0="i+1:word_form_simplified_lowercase" f1="i+2:word_form_simplified_lowercase"/> <feature f0="i-1:word_form_simplified_lowercase" f1="i+1:word_form_simplified_lowercase"/> <feature f0="i-2:part_of_speech_tag" f1="i-1:part_of_speech_tag"/> <feature f0="i-1:part_of_speech_tag" f1="i+1:ambiguity_classes"/> <feature f0="i+1:ambiguity_classes" f1="i+2:ambiguity_classes"/> <!-- 3-gram features --> <feature f0="i-2:part_of_speech_tag" f1="i-1:part_of_speech_tag" f2="i:ambiguity_classes"/> <feature f0="i-1:part_of_speech_tag" f1="i:ambiguity_classes" f2="i+1:ambiguity_classes"/> <feature f0="i-2:part_of_speech_tag" f1="i-1:part_of_speech_tag" f2="i+1:ambiguity_classes"/> <feature f0="i-1:part_of_speech_tag" f1="i+1:ambiguity_classes" f2="i+2:ambiguity_classes"/> <!-- affix features --> <feature f0="i:prefix:2"/> <feature f0="i:prefix:3"/> <feature f0="i:suffix:1"/> <feature f0="i:suffix:2"/> <feature f0="i:suffix:3"/> <feature f0="i:suffix:4"/> <!-- orthographic features --> <feature set="true" f0="i:orthographic"/> <!-- positional features --> <feature set="true" f0="i:positional"/> <!-- word cluster features --> <feature set="true" f0="i-1:word_clusters"/> <feature set="true" f0="i:word_clusters"/> <feature set="true" f0="i+1:word_clusters"/> </feature_template> </configuration>
© 2015 - 2024 Weber Informatics LLC | Privacy Policy