|
||||||||||
| PREV PACKAGE NEXT PACKAGE | FRAMES NO FRAMES | |||||||||
See:
Description
| Class Summary | |
| AmbiguityClass | This class is not integrated yet into the tagger. |
| AmbiguityClasses | This class is not integrated yet into the tagger. |
| CollectionTaggerOutputs | This class will just hold an array of the outputs of all available taggers. |
| ConfMatUnconsidered | Builds a confusion matrix of the confusions which were not even possible to not make, because the correct tag was not in the list of tags that were tried. |
| ConfusionMatrix | |
| ConfusionMatrixUnKnown | |
| CountWrapper | |
| DataWordTag | |
| Dictionary | |
| ErrorTags | |
| Extractor | This class serves as the base class for classes which extract relevant information from a history to give it to the features. |
| ExtractorAllTaggerOutputs | |
| ExtractorFollowing2WClass | |
| ExtractorFollowingWClass | |
| ExtractorFrames | This is an important class for the tagger. |
| ExtractorFramesRare | |
| ExtractorLastVerb | |
| ExtractorLVdist | |
| ExtractorMorpho | |
| ExtractorOutputTag | |
| ExtractorParticles | |
| ExtractorParticlesChris | A class that detects and provides features for common verb particle pairs. |
| Extractors | |
| FeatureIndex | |
| FeatureKey | |
| FeatureTemplate | |
| FeatureTemplates | |
| GlobalHolder | Title: StanfordMaxEnt package. |
| History | |
| HistoryTable | |
| LambdaSolveTagger | This module does the working out of lambda parameters for binary tagger features. |
| MainFrame | |
| MaxentTagger | This is the maximum entropy tagger class has methods for training, saving, tagging sentences etc. |
| Morpho | |
| OutputTags | |
| PairsHolder | |
| ReadData | |
| ReadDataTagged | Reads tagged data from a file and creates a dictionary |
| StaticStringTagger | Static version of StringTagger. |
| StringTagger | Convenience class that takes a String input and outputs the tagged version of the String. |
| TaggerExperiments | This class represents the training samples. |
| TaggerFeature | Holds a Tagger Feature for the loglinear model. |
| TaggerFeatures | This class conatins POS tagger specific features. |
| TaggerOutputHolder | |
| TagInference | |
| TemplateHash | |
| Test | This class is used to run the learned POS tagging model over test data. |
| TestClassifier | |
| TestForwardsBackWards | |
| TestSentence | |
| TestSentenceTagged | |
| TopMistagged | |
| Train | This class is used to train a POS tagger. |
| TTags | This class holds the POS tags, assigns them unique Id-s, knows which tags are open versus closed class, etc. |
| WordTag | |
Maximum Entropy Part-Of-Speech Tagger.
Commands to train tagger: cd /u/nlp/java/samples java -server Train data/dataFile -o $place A tagger will be trained using the data in dataFile. It has to be in the format one sentence per line, where each sentence is represented by [word_tag]+ . By default, the parameters are estimated using conjugate gradient. And test tagger with: java -server -mx256m edu.stanford.nlp.maxent.tagger.Test trainFileRoot testFile -o $place Here trainFileRoot is the prefix of the parameter files of a saved tagger. For example, if the parameter files look like R.holder.X , then trainFileRoot = R . testFile should be in the same format as above for the training data. The easiest way to use the tagger in other applications is via the StaticStringTagger class.
|
||||||||||
| PREV PACKAGE NEXT PACKAGE | FRAMES NO FRAMES | |||||||||