convlab2.policy.mle.crosswoz package

Submodules

convlab2.policy.mle.crosswoz.evaluate module

convlab2.policy.mle.crosswoz.evaluate.calculateF1(predict_golden)
convlab2.policy.mle.crosswoz.evaluate.da_evaluate_simulation(policy)
convlab2.policy.mle.crosswoz.evaluate.end2end_evaluate_simulation(policy)
convlab2.policy.mle.crosswoz.evaluate.evaluate_corpus_f1(policy, data, goal_type=None)
convlab2.policy.mle.crosswoz.evaluate.read_zipped_json(filepath, filename)

convlab2.policy.mle.crosswoz.loader module

class convlab2.policy.mle.crosswoz.loader.Dataset(s_s, a_s)

Bases: torch.utils.data.dataset.Dataset

class convlab2.policy.mle.crosswoz.loader.PolicyDataLoaderCrossWoz

Bases: object

create_dataset(part, batchsz)

convlab2.policy.mle.crosswoz.mle module

class convlab2.policy.mle.crosswoz.mle.MLE(archive_file='/home/travis/build/thu-coai/ConvLab-2/convlab2/policy/mle/crosswoz/models/mle_policy_crosswoz.zip', model_file='https://convlab.blob.core.windows.net/convlab-2/mle_policy_crosswoz.zip')

Bases: convlab2.policy.mle.mle.MLEAbstract

convlab2.policy.mle.crosswoz.train module

class convlab2.policy.mle.crosswoz.train.MLE_Trainer(manager, cfg)

Bases: object

imit_test(epoch, best)

provide an unbiased evaluation of the policy fit on the training dataset

imitating(epoch)

pretrain the policy by simple imitation learning (behavioral cloning)

load(filename='save/best')
policy_loop(data)
save(directory, epoch)
test()

Module contents