opennlp-tools/src/main/java/opennlp/tools/formats/masc/MascDocument.java (5 lines): - line 93: //TODO: make the annotations immutable - line 94: //TODO: should we cleanup the document (e.g. remove sentences without tokens?) - line 166: //TODO: can we know a priori, if we need this filtering? - line 261: // TODO: is it okay that quarks can cross sentence boundary? What are the implications? - line 274: // TODO: can this be translated into Span's methods .crosses()/.contains()? opennlp-tools/src/main/java/opennlp/tools/parser/AbstractContextGenerator.java (5 lines): - line 112: //TODO: make sure multiple punctuation comes out the same - line 129: if (c0.index == 0) { //TODO look at removing case - line 134: if (c1.index == 0) { //TODO look at removing case - line 181: //TODO consider changing - line 195: //TODO consider changing opennlp-docs/src/docbkx/parser.xml (4 lines): - line 123: TODO: Extend this section with more information about the Parse object. - line 132: (TODO: Add a section which explains the two different approaches) - line 146: is doing this to achieve a better performance. (TODO: On which data is the model on - line 155: To train the parser a head rules file is also needed. (TODO: Add documentation about the head rules file) opennlp-tools/src/main/java/opennlp/tools/util/model/BaseModel.java (3 lines): - line 53: // TODO: Provide subclasses access to serializers already in constructor - line 237: // TODO: Discuss this solution, the buffering should - line 257: // TODO: Probably better to use the serializer here directly! opennlp-tools/src/main/java/opennlp/tools/cmdline/entitylinker/EntityLinkerTool.java (3 lines): - line 58: // TODO: Ask Mark if we can remove the type, the user knows upfront if s/he tries - line 65: // TODO: Entity Linker Properties constructor should not duplicate code - line 74: // TODO: It should not just throw Exception. opennlp-tools/src/main/java/opennlp/tools/namefind/TokenNameFinderModel.java (3 lines): - line 48: // TODO: Fix the model validation, on loading via constructors and input streams - line 248: // TODO: Add checks to not put resources where no serializer exists, - line 320: // TODO: Not so nice, because code cannot really be reused by the other create serializer method opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerFactory.java (2 lines): - line 91: // TODO: This could be made functional by creating some default feature generation - line 313: // TODO: This should not be done anymore for 8 models, they can just opennlp-tools/src/main/java/opennlp/tools/namefind/NameFinderME.java (2 lines): - line 87: // TODO: We should deprecate this. And come up with a better solution! - line 234: } // TODO: Maybe it is not a good idea, that these two don't use the context generator ?! opennlp-tools/src/main/java/opennlp/tools/ml/TrainerFactory.java (2 lines): - line 222: // TODO: Need to validate all parameters correctly ... error prone?! - line 246: // TODO: Check data indexing ... opennlp-dl/src/main/java/opennlp/dl/namefinder/NameFinderDL.java (2 lines): - line 168: // TODO: Need to make sure this value is between 0 and 1? - line 238: // TODO: This check should not be needed because the span was found. opennlp-tools/src/main/java/opennlp/tools/formats/brat/BratNameSampleStreamFactory.java (2 lines): - line 121: // TODO: Add an optional parameter to search recursive - line 122: // TODO: How to handle the error here ? terminate the tool? not nice if used by API! opennlp-tools/src/main/java/opennlp/tools/ml/maxent/GISTrainer.java (2 lines): - line 598: // TODO: We got interrupted, but that is currently not really supported! - line 679: // TODO: check interruption status here, if interrupted set a poisoned flag and return opennlp-tools/src/main/java/opennlp/tools/formats/ConllXSentenceSampleStreamFactory.java (1 line): - line 41: // TODO: make chunk size configurable opennlp-tools/src/main/java/opennlp/tools/parser/Parse.java (1 line): - line 467: // TODO: Use the commented code in next bigger release, opennlp-tools/src/main/java/opennlp/tools/postag/POSTaggerME.java (1 line): - line 326: // TODO: This will probably cause issue, since the feature generator uses the outcomes array opennlp-tools/src/main/java/opennlp/tools/lemmatizer/LemmatizerME.java (1 line): - line 278: // TODO: This will probably cause issue, since the feature generator uses the outcomes array opennlp-tools/src/main/java/opennlp/tools/dictionary/serializer/DictionaryEntryPersistor.java (1 line): - line 59: // TODO: should check for invalid format, make it save opennlp-docs/src/docbkx/uima-integration.xml (1 line): - line 102: TODO: Extend this documentation with information about the individual components. opennlp-tools/src/main/java/opennlp/tools/cmdline/parser/BuildModelUpdaterTool.java (1 line): - line 54: // TODO: training individual models should be in the chunking parser, not here opennlp-tools/src/main/java/opennlp/tools/postag/POSModel.java (1 line): - line 76: // TODO: This fails probably for the sequence model ... ?! opennlp-tools/src/main/java/opennlp/tools/ml/model/MaxentModel.java (1 line): - line 79: // TODO: This should be removed, can't be used anyway without format spec opennlp-tools/src/main/java/opennlp/tools/formats/frenchtreebank/ConstitParseSampleStreamFactory.java (1 line): - line 40: // TODO: The parameters have an encoding, but the data is in xml opennlp-tools/src/main/java/opennlp/tools/parser/ParserEventTypeEnum.java (1 line): - line 29: // TODO Add reason why those enum values are deprecated opennlp-tools/src/main/java/opennlp/tools/sentdetect/EmptyLinePreprocessorStream.java (1 line): - line 32: * - TODO: Terminates last document with empty line if it is missing
opennlp-tools/src/main/java/opennlp/tools/namefind/DefaultNameContextGenerator.java (1 line): - line 102: // TODO: These should be moved out here in its own feature generator! opennlp-tools/src/main/java/opennlp/tools/postag/POSDictionary.java (1 line): - line 228: // TODO: The dictionary API needs to be improved to do this better! opennlp-tools/src/main/java/opennlp/tools/cmdline/languagemodel/NGramLanguageModelTool.java (1 line): - line 67: // TODO : use a Tokenizer here opennlp-tools/src/main/java/opennlp/tools/parser/AbstractBottomUpParser.java (1 line): - line 287: //TODO: Need to look at this for K-best parsing cases opennlp-tools/src/main/java/opennlp/tools/lemmatizer/DefaultLemmatizerSequenceValidator.java (1 line): - line 27: //TODO implement this opennlp-tools/src/main/java/opennlp/tools/cmdline/parser/CheckModelUpdaterTool.java (1 line): - line 58: // TODO: Maybe that should be part of the ChunkingParser ... opennlp-tools/src/main/java/opennlp/tools/formats/ad/ADSentenceSampleStream.java (1 line): - line 138: // TODO Merge these patterns opennlp-tools/src/main/java/opennlp/tools/lemmatizer/LemmaSampleSequenceStream.java (1 line): - line 67: // TODO: Should be implemented for Perceptron sequence learning ... opennlp-tools/src/main/java/opennlp/tools/tokenize/WordpieceTokenizer.java (1 line): - line 80: // TODO: Implement this. opennlp-tools/src/main/java/opennlp/tools/formats/Conll03NameSampleStreamFactory.java (1 line): - line 59: // TODO: support the other languages with this CoNLL. opennlp-tools/src/main/java/opennlp/tools/langdetect/LanguageDetectorME.java (1 line): - line 181: // TODO -- once OPENNLP-1261 is fixed, opennlp-uima/src/main/java/opennlp/uima/parser/Parser.java (1 line): - line 185: // TODO: We need a strategy to handle the case that a full opennlp-uima/src/main/java/opennlp/uima/namefind/NameFinder.java (1 line): - line 184: // TODO: Create confidence FS opennlp-tools/src/main/java/opennlp/tools/parser/ParserModel.java (1 line): - line 278: // TODO: (All!) Update model methods should make sure properties are copied correctly ... opennlp-tools/src/main/java/opennlp/tools/cmdline/StreamFactoryRegistry.java (1 line): - line 227: // TODO: Need to check if it can produce the desired output opennlp-tools/src/main/java/opennlp/tools/cmdline/parser/ParserTrainerTool.java (1 line): - line 122: // TODO: Add param to train tree insert parser opennlp-tools/src/main/java/opennlp/tools/formats/letsmt/LetsmtDocument.java (1 line): - line 96: // TODO: The sentence should contain the id, so it can be tracked back to the opennlp-tools/src/main/java/opennlp/tools/chunker/ChunkerME.java (1 line): - line 173: // TODO: This will probably cause issue, since the feature generator uses the outcomes array opennlp-tools/src/main/java/opennlp/tools/util/featuregen/InSpanGenerator.java (1 line): - line 77: // TODO: consider generation start and continuation features opennlp-tools/src/main/java/opennlp/tools/formats/masc/MascSentence.java (1 line): - line 60: // TODO: evaluate the necessity: HashMaps are O(1), right? opennlp-tools/src/main/java/opennlp/tools/formats/brat/BratDocumentParser.java (1 line): - line 101: // TODO: Token breaks should be enforced on name span boundaries opennlp-tools/src/main/java/opennlp/tools/sentdetect/SentenceDetectorME.java (1 line): - line 355: // TODO: Fix the EventStream to throw exceptions when training goes wrong opennlp-docs/src/docbkx/coref.xml (1 line): - line 34: TODO: Write more documentation about the coref component. Any contributions opennlp-tools/src/main/java/opennlp/tools/formats/letsmt/LetsmtSentenceStreamFactory.java (1 line): - line 73: // TODO Implement a filter stream to remove splits which are not at an eos char opennlp-tools/src/main/java/opennlp/tools/parser/treeinsert/ParserEventStream.java (1 line): - line 204: //TODO: Consider whether we need to set this label or train parses at all. opennlp-tools/src/main/java/opennlp/tools/formats/masc/MascNamedEntitySampleStream.java (1 line): - line 84: // TODO: should the user decide about clearAdaptiveData? opennlp-tools/src/main/java/opennlp/tools/chunker/ChunkSampleSequenceStream.java (1 line): - line 73: // TODO: Should be implemented for Perceptron sequence learning ... opennlp-tools/src/main/java/opennlp/tools/cmdline/namefind/TokenNameFinderTrainerTool.java (1 line): - line 132: // TODO: Support Custom resources: opennlp-tools/src/main/java/opennlp/tools/namefind/NameSample.java (1 line): - line 283: // TODO: Should throw another exception, and then convert it into an IOException in the stream opennlp-tools/src/main/java/opennlp/tools/sentdetect/SentenceSample.java (1 line): - line 94: // TODO: This one must output the tags! opennlp-tools/src/main/java/opennlp/tools/ml/AbstractTrainer.java (1 line): - line 101: // TODO: Need to validate all parameters correctly ... error prone?! opennlp-tools/src/main/java/opennlp/tools/ml/perceptron/PerceptronTrainer.java (1 line): - line 370: // TODO: Make averaging configurable !!!