def from_params(cls, vocab, params: Params) -> 'WikiTablesMmlSemanticParser':
question_embedder = TextFieldEmbedder.from_params(vocab, params.pop("question_embedder"))
action_embedding_dim = params.pop_int("action_embedding_dim")
encoder = Seq2SeqEncoder.from_params(params.pop("encoder"))
entity_encoder = Seq2VecEncoder.from_params(params.pop('entity_encoder'))
max_decoding_steps = params.pop_int("max_decoding_steps")
mixture_feedforward_type = params.pop('mixture_feedforward', None)
if mixture_feedforward_type is not None:
mixture_feedforward = FeedForward.from_params(mixture_feedforward_type)
else:
mixture_feedforward = None
decoder_beam_search = BeamSearch.from_params(params.pop("decoder_beam_search"))
input_attention = Attention.from_params(params.pop("attention"))
training_beam_size = params.pop_int('training_beam_size', None)
use_neighbor_similarity_for_linking = params.pop_bool('use_neighbor_similarity_for_linking', False)
dropout = params.pop_float('dropout', 0.0)
num_linking_features = params.pop_int('num_linking_features', 10)
tables_directory = params.pop('tables_directory', '/wikitables/')
rule_namespace = params.pop('rule_namespace', 'rule_labels')
params.assert_empty(cls.__name__)
return cls(vocab,
question_embedder=question_embedder,
action_embedding_dim=action_embedding_dim,
encoder=encoder,
entity_encoder=entity_encoder,
mixture_feedforward=mixture_feedforward,
decoder_beam_search=decoder_beam_search,
max_decoding_steps=max_decoding_steps,
input_attention=input_attention,
training_beam_size=training_beam_size,
use_neighbor_similarity_for_linking=use_neighbor_similarity_for_linking,
dropout=dropout,
num_linking_features=num_linking_features,
tables_directory=tables_directory,
rule_namespace=rule_namespace)
def from_params(cls, vocab,
params: Params) -> 'WikiTablesErmSemanticParser':
question_embedder = TextFieldEmbedder.from_params(
vocab, params.pop("question_embedder"))
action_embedding_dim = params.pop_int("action_embedding_dim")
encoder = Seq2SeqEncoder.from_params(params.pop("encoder"))
entity_encoder = Seq2VecEncoder.from_params(
params.pop('entity_encoder'))
mixture_feedforward_type = params.pop('mixture_feedforward', None)
if mixture_feedforward_type is not None:
mixture_feedforward = FeedForward.from_params(
mixture_feedforward_type)
else:
mixture_feedforward = None
input_attention = Attention.from_params(params.pop("attention"))
decoder_beam_size = params.pop_int("decoder_beam_size")
decoder_num_finished_states = params.pop_int(
"decoder_num_finished_states", None)
max_decoding_steps = params.pop_int("max_decoding_steps")
normalize_beam_score_by_length = params.pop(
"normalize_beam_score_by_length", False)
use_neighbor_similarity_for_linking = params.pop_bool(
"use_neighbor_similarity_for_linking", False)
dropout = params.pop_float('dropout', 0.0)
num_linking_features = params.pop_int('num_linking_features', 10)
tables_directory = params.pop('tables_directory', '/wikitables/')
rule_namespace = params.pop('rule_namespace', 'rule_labels')
checklist_cost_weight = params.pop_float("checklist_cost_weight", 0.6)
mml_model_file = params.pop('mml_model_file', None)
params.assert_empty(cls.__name__)
return cls(
vocab,
question_embedder=question_embedder,
action_embedding_dim=action_embedding_dim,
encoder=encoder,
entity_encoder=entity_encoder,
mixture_feedforward=mixture_feedforward,
input_attention=input_attention,
decoder_beam_size=decoder_beam_size,
decoder_num_finished_states=decoder_num_finished_states,
max_decoding_steps=max_decoding_steps,
normalize_beam_score_by_length=normalize_beam_score_by_length,
checklist_cost_weight=checklist_cost_weight,
use_neighbor_similarity_for_linking=
use_neighbor_similarity_for_linking,
dropout=dropout,
num_linking_features=num_linking_features,
tables_directory=tables_directory,
rule_namespace=rule_namespace,
initial_mml_model_file=mml_model_file)
def from_params(cls, vocab, params: Params) -> 'NlvrDirectSemanticParser':
sentence_embedder_params = params.pop("sentence_embedder")
sentence_embedder = TextFieldEmbedder.from_params(
vocab, sentence_embedder_params)
action_embedding_dim = params.pop_int('action_embedding_dim')
encoder = Seq2SeqEncoder.from_params(params.pop("encoder"))
dropout = params.pop_float('dropout', 0.0)
input_attention = Attention.from_params(params.pop("attention"))
decoder_beam_search = BeamSearch.from_params(
params.pop("decoder_beam_search"))
max_decoding_steps = params.pop_int("max_decoding_steps")
params.assert_empty(cls.__name__)
return cls(vocab,
sentence_embedder=sentence_embedder,
action_embedding_dim=action_embedding_dim,
encoder=encoder,
input_attention=input_attention,
decoder_beam_search=decoder_beam_search,
max_decoding_steps=max_decoding_steps,
dropout=dropout)
def from_params(cls, vocab,
params: Params) -> 'NlvrCoverageSemanticParser':
sentence_embedder_params = params.pop("sentence_embedder")
sentence_embedder = TextFieldEmbedder.from_params(
vocab, sentence_embedder_params)
action_embedding_dim = params.pop_int('action_embedding_dim')
encoder = Seq2SeqEncoder.from_params(params.pop("encoder"))
dropout = params.pop_float('dropout', 0.0)
input_attention = Attention.from_params(params.pop("attention"))
beam_size = params.pop_int('beam_size')
max_num_finished_states = params.pop_int('max_num_finished_states',
None)
normalize_beam_score_by_length = params.pop_bool(
'normalize_beam_score_by_length', False)
max_decoding_steps = params.pop_int("max_decoding_steps")
checklist_cost_weight = params.pop_float("checklist_cost_weight", 0.6)
dynamic_cost_weight = params.pop("dynamic_cost_weight", None)
penalize_non_agenda_actions = params.pop_bool(
"penalize_non_agenda_actions", False)
initial_mml_model_file = params.pop("initial_mml_model_file", None)
params.assert_empty(cls.__name__)
return cls(
vocab,
sentence_embedder=sentence_embedder,
action_embedding_dim=action_embedding_dim,
encoder=encoder,
input_attention=input_attention,
beam_size=beam_size,
max_num_finished_states=max_num_finished_states,
dropout=dropout,
max_decoding_steps=max_decoding_steps,
normalize_beam_score_by_length=normalize_beam_score_by_length,
checklist_cost_weight=checklist_cost_weight,
dynamic_cost_weight=dynamic_cost_weight,
penalize_non_agenda_actions=penalize_non_agenda_actions,
initial_mml_model_file=initial_mml_model_file)
def test_can_build_from_params(self):
params = Params({'similarity_function': {'type': 'cosine'}, 'normalize': False})
attention = Attention.from_params(params)
# pylint: disable=protected-access
assert attention._similarity_function.__class__.__name__ == 'CosineSimilarity'
assert attention._normalize is False
def from_params(cls, vocab: Vocabulary, params: Params) -> 'MtGan': # type: ignore
# pylint: disable=arguments-differ
vocab_namespace_A = params.pop("vocab_namespace_A", "vocab_A")
vocab_namespace_B = params.pop("vocab_namespace_B", "vocab_B")
num_classes_A = vocab.get_vocab_size(namespace=vocab_namespace_A)
num_classes_B = vocab.get_vocab_size(namespace=vocab_namespace_B)
params_generators = params.pop("generators")
if params_generators.pop("type") == "rnn2rnn":
generators_embedding_dim = params_generators.pop("embedding_dim")
embedding_A_generator = Embedding(num_embeddings=num_classes_A, embedding_dim=generators_embedding_dim)
embedding_B_generator = Embedding(num_embeddings=num_classes_B, embedding_dim=generators_embedding_dim)
params_encoder_generators = params_generators.pop("encoder")
generator_A_to_B_encoder = Seq2SeqEncoder.from_params(params_encoder_generators.duplicate())
generator_B_to_A_encoder = Seq2SeqEncoder.from_params(params_encoder_generators.duplicate())
generator_attention_params = params_generators.pop("attention")
attention_generator_A_to_B = Attention.from_params(generator_attention_params.duplicate())
attention_generator_B_to_A = Attention.from_params(generator_attention_params.duplicate())
generators_max_decoding_steps = params_generators.pop("max_decoding_steps")
generator_A_to_B = Rnn2Rnn(vocab=vocab,
source_embedding=embedding_A_generator,
target_embedding=embedding_B_generator,
encoder=generator_A_to_B_encoder,
max_decoding_steps=generators_max_decoding_steps,
target_namespace=vocab_namespace_B,
attention=attention_generator_A_to_B)
generator_B_to_A = Rnn2Rnn(vocab=vocab,
source_embedding=embedding_B_generator,
target_embedding=embedding_A_generator,
encoder=generator_B_to_A_encoder,
max_decoding_steps=generators_max_decoding_steps,
target_namespace=vocab_namespace_A,
attention=attention_generator_B_to_A)
else:
raise ConfigurationError(message="This generators model type is not supported")
discriminators_params = params.pop("discriminators")
if discriminators_params.pop("type") == "seq2prob":
params_encoder_discriminators = discriminators_params.pop("encoder")
discriminator_A_encoder = Seq2VecEncoder.from_params(params_encoder_discriminators.duplicate())
discriminator_B_encoder = Seq2VecEncoder.from_params(params_encoder_discriminators.duplicate())
discriminators_embedding_dim = discriminators_params.pop("embedding_dim")
embedding_A_discriminator = Embedding(num_classes_A, discriminators_embedding_dim)
embedding_B_discriminator = Embedding(num_classes_B, discriminators_embedding_dim)
discriminator_A = Seq2Prob(vocab=vocab, encoder=discriminator_A_encoder, embedding=embedding_A_discriminator)
discriminator_B = Seq2Prob(vocab=vocab, encoder=discriminator_B_encoder, embedding=embedding_B_discriminator)
else:
raise ConfigurationError(message="This discriminators model type is not supported")
return cls(vocab=vocab,
generator_A_to_B=generator_A_to_B,
generator_B_to_A=generator_B_to_A,
discriminator_A=discriminator_A,
discriminator_B=discriminator_B,
vocab_namespace_A=vocab_namespace_A,
vocab_namespace_B=vocab_namespace_B)
