convlab2.nlg.template.camrest package

Submodules

convlab2.nlg.template.camrest.evaluate module

Evaluate NLG models on utterances of Multiwoz test dataset Metric: dataset level BLEU-4, slot error rate Usage: python evaluate.py [usr|sys|all]

convlab2.nlg.template.camrest.evaluate.get_bleu4(dialog_acts, golden_utts, gen_utts)

convlab2.nlg.template.camrest.nlg module

class convlab2.nlg.template.camrest.nlg.TemplateNLG(is_user, mode='manual')

Bases: convlab2.nlg.nlg.NLG

generate(dialog_acts)

NLG for Multiwoz dataset

Args:
dialog_acts:

{da1:[[slot1,value1],…], da2:…}

Returns:

generated sentence

static read_json(filename)
convlab2.nlg.template.camrest.nlg.example()
convlab2.nlg.template.camrest.nlg.test()

Module contents