cogcomp-nlp
cogcomp-nlp copied to clipboard
CogComp
CogComp's Natural Language Processing Libraries and Demos: Modules include lemmatizer, ner, pos, prep-srl, quantifier, question type, relation-extraction, similarity, temporal normalizer, tokenizer, t...
Use Maven Dependency Analyzer to find the dependencies not needed and drop them: https://maven.apache.org/shared/maven-dependency-analyzer/
I experienced that our latest tokenizer breaks on empty string "" and "\n" at the least. It produces an 'ArrayOutOfBound Exception'
``` Fifty-seven chapels are located throughout the campus. ```  FYI @qiangning
We have an explicit sentence view: make sure it is written out when serialized, and that it is read when deserialized. Remove TextAnnotation.sentences as Sentence data structure creates potential inconsistencies...
Why there are two "ViewName"s in jsons? Seems redundant ...
For this input: ``` On each chromosome pair are smaller bodies (genes) which control all aspects of a bird, its colour, shape, bone structure, length of feather, size etc. The...
The POSAnnotator on giving an empty string as input, throws an error. Ideally it should not do that. Exception in thread "main" java.lang.IndexOutOfBoundsException: Index: 0, Size: 0
Looking at code to try to understand how to use POSAnnotator I see an opportunity for a minor improvement. In https://github.com/CogComp/cogcomp-nlp/blob/49f226331c203ba3ff4f20b51a1a4c97bce999b7/pos/src/main/java/edu/illinois/cs/cogcomp/pos/POSAnnotator.java#L94 and line 97 the code seems to be doing...
The test is failing (its handled though): ``` 07:24:42 ERROR TestWriteSVMLightFormat:122 - Expected exception is caught: java.lang.IllegalArgumentException: ERROR: edu.illinois.cs.cogcomp.edison.utilities.WriteSVMLightFormat constructed expecting binary classification task. Label must be either 'true' or...
Ported most functionalities, but there is still an issue when loading resource file from datastore, will fix it soon.
