krickert opened a new issue, #1469: URL: https://github.com/apache/opennlp/issues/1469
#1323 adds `FeedforwardDependencyParser`, `FeedforwardDependencyTrainer` and `FeedforwardDependencyModel`, but the command line only covers `DependencyParserME`. Users who want to train or run the neural parser have to write Java. This issue adds the matching tools in `opennlp-cli`, mirrored from the existing `depparse` tools: - `FeedforwardDependencyParserTrainerTool`: trains from the same sample formats as `DependencyParserTrainerTool`, takes the trainer's `TrainingParameters` keys, and can refine an existing model with a beam. - `FeedforwardDependencyParserTool`: parses tagged sentences from stdin with a model file, with a `-beamSize` option. - `FeedforwardDependencyParserEvaluatorTool`: scores a model on a treebank with `DependencyEvaluator`, with the same output as the ME evaluator. - A `FeedforwardDependencyModelLoader` for the tools above. The manual's dependency parser chapter gets a command-line subsection for the new tools. A cross-validator tool is out of scope until a `FeedforwardDependencyCrossValidator` exists. Depends on #1323 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
