convlab2.nlu package¶
Subpackages¶
- convlab2.nlu.jointBERT package
- Subpackages
- Submodules
- convlab2.nlu.jointBERT.dataloader module
- convlab2.nlu.jointBERT.jointBERT module
- convlab2.nlu.jointBERT.test module
- convlab2.nlu.jointBERT.train module
- Module contents
- convlab2.nlu.milu package
- convlab2.nlu.svm package
- Subpackages
- Submodules
- convlab2.nlu.svm.Classifier module
- convlab2.nlu.svm.Features module
- convlab2.nlu.svm.Tuples module
- convlab2.nlu.svm.dataset_walker module
- convlab2.nlu.svm.sutils module
- convlab2.nlu.svm.train module
- Module contents
Submodules¶
convlab2.nlu.evaluate module¶
Evaluate NLU models on specified dataset Metric: dataset level Precision/Recall/F1 Usage: python evaluate.py [CrossWOZ|MultiWOZ|Camrest] [BERTNLU|MILU|SVMNLU] [usr|sys|all]
-
convlab2.nlu.evaluate.
calculateF1
(predict_golden)¶
convlab2.nlu.nlu module¶
Natural Language Understanding Interface.
-
class
convlab2.nlu.nlu.
NLU
¶ Bases:
convlab2.util.module.Module
NLU module interface.
-
predict
(utterance, context=[])¶ Predict the dialog act of a natural language utterance.
- Args:
- utterance (str):
A natural language utterance.
- context (list of str):
Previous utterances.
- Returns:
- action (list of list):
The dialog act of utterance.
-