def setup_method(self):
EvalbBracketingScorer.compile_evalb()
super().setup_method()
self.set_up_model(
FIXTURES_ROOT / "syntax" / "constituency_parser" / "constituency_parser.json",
FIXTURES_ROOT / "syntax" / "example_ptb.trees",
)
def setUp(self):
EvalbBracketingScorer.compile_evalb()
super(SpanConstituencyParserTest, self).setUp()
self.set_up_model(
self.FIXTURES_ROOT / u"constituency_parser" /
u"constituency_parser.json",
self.FIXTURES_ROOT / u"data" / u"example_ptb.trees")
def __init__(
self,
vocab: Vocabulary,
text_field_embedder: TextFieldEmbedder,
span_extractor: SpanExtractor,
encoder: Seq2SeqEncoder,
feedforward: FeedForward = None,
pos_tag_embedding: Embedding = None,
initializer: InitializerApplicator = InitializerApplicator(),
regularizer: Optional[RegularizerApplicator] = None,
evalb_directory_path: str = DEFAULT_EVALB_DIR,
) -> None:
super().__init__(vocab, regularizer)
self.text_field_embedder = text_field_embedder
self.span_extractor = span_extractor
self.num_classes = self.vocab.get_vocab_size("labels")
self.encoder = encoder
self.feedforward_layer = TimeDistributed(
feedforward) if feedforward else None
self.pos_tag_embedding = pos_tag_embedding or None
if feedforward is not None:
output_dim = feedforward.get_output_dim()
else:
output_dim = span_extractor.get_output_dim()
self.tag_projection_layer = TimeDistributed(
Linear(output_dim, self.num_classes))
representation_dim = text_field_embedder.get_output_dim()
if pos_tag_embedding is not None:
representation_dim += pos_tag_embedding.get_output_dim()
check_dimensions_match(
representation_dim,
encoder.get_input_dim(),
"representation dim (tokens + optional POS tags)",
"encoder input dim",
)
check_dimensions_match(
encoder.get_output_dim(),
span_extractor.get_input_dim(),
"encoder input dim",
"span extractor input dim",
)
if feedforward is not None:
check_dimensions_match(
span_extractor.get_output_dim(),
feedforward.get_input_dim(),
"span extractor output dim",
"feedforward input dim",
)
self.tag_accuracy = CategoricalAccuracy()
if evalb_directory_path is not None:
self._evalb_score = EvalbBracketingScorer(evalb_directory_path)
else:
self._evalb_score = None
initializer(self)
def test_evalb_correctly_calculates_bracketing_metrics_over_multiple_trees(self):
tree1 = Tree.fromstring("(S (VP (D the) (NP dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1, tree2], [tree2, tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 0.875
assert metrics["evalb_precision"] == 0.875
assert metrics["evalb_f1_measure"] == 0.875
def test_evalb_correctly_scores_identical_trees(self):
tree1 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1], [tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 1.0
assert metrics["evalb_precision"] == 1.0
assert metrics["evalb_f1_measure"] == 1.0
def test_evalb_correctly_calculates_bracketing_metrics_over_multiple_trees(self):
tree1 = Tree.fromstring("(S (VP (D the) (NP dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1, tree2], [tree2, tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 0.875
assert metrics["evalb_precision"] == 0.875
assert metrics["evalb_f1_measure"] == 0.875
def test_evalb_correctly_scores_identical_trees(self):
tree1 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1], [tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 1.0
assert metrics["evalb_precision"] == 1.0
assert metrics["evalb_f1_measure"] == 1.0
def test_evalb_correctly_scores_imperfect_trees(self):
# Change to constiutency label (VP ... )should effect scores, but change to POS
# tag (NP dog) should have no effect.
tree1 = Tree.fromstring("(S (VP (D the) (NP dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1], [tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 0.75
assert metrics["evalb_precision"] == 0.75
assert metrics["evalb_f1_measure"] == 0.75
def test_evalb_correctly_scores_imperfect_trees(self):
# Change to constiutency label (VP ... )should effect scores, but change to POS
# tag (NP dog) should have no effect.
tree1 = Tree.fromstring("(S (VP (D the) (NP dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1], [tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 0.75
assert metrics["evalb_precision"] == 0.75
assert metrics["evalb_f1_measure"] == 0.75
def test_evalb_with_terrible_trees_handles_nan_f1(self):
# If precision and recall are zero, evalb returns nan f1.
# This checks that we handle the zero division.
tree1 = Tree.fromstring(u"(PP (VROOT (PP That) (VROOT (PP could) "
u"(VROOT (PP cost) (VROOT (PP him))))) (PP .))")
tree2 = Tree.fromstring(u"(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1], [tree2])
metrics = evalb_scorer.get_metric()
assert metrics[u"evalb_recall"] == 0.0
assert metrics[u"evalb_precision"] == 0.0
assert metrics[u"evalb_f1_measure"] == 0.0
def test_evalb_with_terrible_trees_handles_nan_f1(self):
# If precision and recall are zero, evalb returns nan f1.
# This checks that we handle the zero division.
tree1 = Tree.fromstring("(PP (VROOT (PP That) (VROOT (PP could) "
"(VROOT (PP cost) (VROOT (PP him))))) (PP .))")
tree2 = Tree.fromstring("(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
evalb_scorer = EvalbBracketingScorer()
evalb_scorer([tree1], [tree2])
metrics = evalb_scorer.get_metric()
assert metrics["evalb_recall"] == 0.0
assert metrics["evalb_precision"] == 0.0
assert metrics["evalb_f1_measure"] == 0.0
def __init__(self,
vocab: Vocabulary,
text_field_embedder: TextFieldEmbedder,
span_extractor: SpanExtractor,
encoder: Seq2SeqEncoder,
feedforward_layer: FeedForward = None,
initializer: InitializerApplicator = InitializerApplicator(),
regularizer: Optional[RegularizerApplicator] = None,
evalb_directory_path: str = None) -> None:
super(SpanConstituencyParser, self).__init__(vocab, regularizer)
self.text_field_embedder = text_field_embedder
self.span_extractor = span_extractor
self.num_classes = self.vocab.get_vocab_size("labels")
self.encoder = encoder
self.feedforward_layer = TimeDistributed(
feedforward_layer) if feedforward_layer else None
if feedforward_layer is not None:
output_dim = feedforward_layer.get_output_dim()
else:
output_dim = span_extractor.get_output_dim()
self.tag_projection_layer = TimeDistributed(
Linear(output_dim, self.num_classes))
check_dimensions_match(text_field_embedder.get_output_dim(),
encoder.get_input_dim(),
"text field embedding dim", "encoder input dim")
if feedforward_layer is not None:
check_dimensions_match(encoder.get_output_dim(),
feedforward_layer.get_input_dim(),
"stacked encoder output dim",
"feedforward input dim")
self.metrics = {
label: F1Measure(index)
for index, label in self.vocab.get_index_to_token_vocabulary(
"labels").items()
}
if evalb_directory_path is not None:
self._evalb_score = EvalbBracketingScorer(evalb_directory_path)
else:
self._evalb_score = None
initializer(self)
def __init__(self,
vocab: Vocabulary,
text_field_embedder: TextFieldEmbedder,
span_extractor: SpanExtractor,
encoder: Seq2SeqEncoder,
feedforward_layer: FeedForward = None,
pos_tag_embedding: Embedding = None,
initializer: InitializerApplicator = InitializerApplicator(),
regularizer: Optional[RegularizerApplicator] = None,
evalb_directory_path: str = None) -> None:
super(SpanConstituencyParser, self).__init__(vocab, regularizer)
self.text_field_embedder = text_field_embedder
self.span_extractor = span_extractor
self.num_classes = self.vocab.get_vocab_size("labels")
self.encoder = encoder
self.feedforward_layer = TimeDistributed(feedforward_layer) if feedforward_layer else None
self.pos_tag_embedding = pos_tag_embedding or None
if feedforward_layer is not None:
output_dim = feedforward_layer.get_output_dim()
else:
output_dim = span_extractor.get_output_dim()
self.tag_projection_layer = TimeDistributed(Linear(output_dim, self.num_classes))
representation_dim = text_field_embedder.get_output_dim()
if pos_tag_embedding is not None:
representation_dim += pos_tag_embedding.get_output_dim()
check_dimensions_match(representation_dim,
encoder.get_input_dim(),
"representation dim (tokens + optional POS tags)",
"encoder input dim")
check_dimensions_match(encoder.get_output_dim(),
span_extractor.get_input_dim(),
"encoder input dim",
"span extractor input dim")
if feedforward_layer is not None:
check_dimensions_match(span_extractor.get_output_dim(),
feedforward_layer.get_input_dim(),
"span extractor output dim",
"feedforward input dim")
self.tag_accuracy = CategoricalAccuracy()
if evalb_directory_path is not None:
self._evalb_score = EvalbBracketingScorer(evalb_directory_path)
else:
self._evalb_score = None
initializer(self)
def multiple_runs(
global_rank: int,
world_size: int,
gpu_id: Union[int, torch.device],
metric: EvalbBracketingScorer,
metric_kwargs: Dict[str, List[Any]],
desired_values: Dict[str, Any],
exact: Union[bool, Tuple[float, float]] = True,
):
kwargs = {}
# Use the arguments meant for the process with rank `global_rank`.
for argname in metric_kwargs:
kwargs[argname] = metric_kwargs[argname][global_rank]
for i in range(200):
metric(**kwargs)
metric_values = metric.get_metric()
for key in desired_values:
assert desired_values[key] == metric_values[key]
def test_distributed_evalb(self):
tree1 = Tree.fromstring(
"(S (VP (D the) (NP dog)) (VP (V chased) (NP (D the) (N cat))))")
tree2 = Tree.fromstring(
"(S (NP (D the) (N dog)) (VP (V chased) (NP (D the) (N cat))))")
predicted_trees = [[tree1], [tree2]]
gold_trees = [[tree2], [tree2]]
metric_kwargs = {
"predicted_trees": predicted_trees,
"gold_trees": gold_trees
}
desired_values = {
"evalb_recall": 0.875,
"evalb_precision": 0.875,
"evalb_f1_measure": 0.875,
}
run_distributed_test(
[-1, -1],
global_distributed_metric,
EvalbBracketingScorer(),
metric_kwargs,
desired_values,
exact=True,
)
def tearDown(self):
EvalbBracketingScorer.clean_evalb()
super().tearDown()
if __name__ == "__main__":
initial_working_dir = os.getcwd()
module_root = _get_module_root().parent
logger.info("Changing directory to %s", module_root)
os.chdir(module_root)
if len(sys.argv) >= 2 and sys.argv[1].lower() == "gpu":
train_fixture_gpu("allennlp/tests/fixtures/srl/")
else:
models = [
'biaffine_dependency_parser',
'bidaf',
'dialog_qa',
'constituency_parser',
'coref',
'decomposable_attention',
'encoder_decoder/simple_seq2seq',
'semantic_parsing/nlvr_coverage_semantic_parser',
'semantic_parsing/nlvr_direct_semantic_parser',
'semantic_parsing/wikitables',
'semantic_parsing/atis',
'srl',
]
for model in models:
if model == 'constituency_parser':
EvalbBracketingScorer.compile_evalb()
train_fixture(f"allennlp/tests/fixtures/{model}/")
logger.info("Changing directory back to %s", initial_working_dir)
os.chdir(initial_working_dir)
logger.info("Changing directory to %s", module_root)
os.chdir(module_root)
if len(sys.argv) >= 2 and sys.argv[1].lower() == "gpu":
train_fixture_gpu("allennlp/tests/fixtures/srl/")
else:
models = [
'biaffine_dependency_parser',
'bidaf',
'dialog_qa',
'constituency_parser',
'coref',
'decomposable_attention',
'encoder_decoder/composed_seq2seq',
'encoder_decoder/simple_seq2seq',
'encoder_decoder/copynet_seq2seq',
'semantic_parsing/nlvr_coverage_semantic_parser',
'semantic_parsing/nlvr_direct_semantic_parser',
'semantic_parsing/wikitables',
'semantic_parsing/quarel',
'semantic_parsing/quarel/zeroshot',
'semantic_parsing/quarel/tagger',
'semantic_parsing/atis',
'srl',
]
for model in models:
if model == 'constituency_parser':
EvalbBracketingScorer.compile_evalb()
train_fixture(f"allennlp/tests/fixtures/{model}/")
logger.info("Changing directory back to %s", initial_working_dir)
os.chdir(initial_working_dir)
def setup_method(self):
super().setup_method()
EvalbBracketingScorer.compile_evalb()
def setUp(self):
EvalbBracketingScorer.compile_evalb()
super(SpanConstituencyParserTest, self).setUp()
self.set_up_model("tests/fixtures/constituency_parser/constituency_parser.json",
"tests/fixtures/data/example_ptb.trees")
class SpanConstituencyParser(Model):
"""
This `SpanConstituencyParser` simply encodes a sequence of text
with a stacked `Seq2SeqEncoder`, extracts span representations using a
`SpanExtractor`, and then predicts a label for each span in the sequence.
These labels are non-terminal nodes in a constituency parse tree, which we then
greedily reconstruct.
# Parameters
vocab : `Vocabulary`, required
A Vocabulary, required in order to compute sizes for input/output projections.
text_field_embedder : `TextFieldEmbedder`, required
Used to embed the `tokens` `TextField` we get as input to the model.
span_extractor : `SpanExtractor`, required.
The method used to extract the spans from the encoded sequence.
encoder : `Seq2SeqEncoder`, required.
The encoder that we will use in between embedding tokens and
generating span representations.
feedforward : `FeedForward`, required.
The FeedForward layer that we will use in between the encoder and the linear
projection to a distribution over span labels.
pos_tag_embedding : `Embedding`, optional.
Used to embed the `pos_tags` `SequenceLabelField` we get as input to the model.
initializer : `InitializerApplicator`, optional (default=`InitializerApplicator()`)
Used to initialize the model parameters.
evalb_directory_path : `str`, optional (default=`DEFAULT_EVALB_DIR`)
The path to the directory containing the EVALB executable used to score
bracketed parses. By default, will use the EVALB included with allennlp,
which is located at allennlp/tools/EVALB . If `None`, EVALB scoring
is not used.
"""
def __init__(
self,
vocab: Vocabulary,
text_field_embedder: TextFieldEmbedder,
span_extractor: SpanExtractor,
encoder: Seq2SeqEncoder,
feedforward: FeedForward = None,
pos_tag_embedding: Embedding = None,
initializer: InitializerApplicator = InitializerApplicator(),
evalb_directory_path: str = DEFAULT_EVALB_DIR,
**kwargs,
) -> None:
super().__init__(vocab, **kwargs)
self.text_field_embedder = text_field_embedder
self.span_extractor = span_extractor
self.num_classes = self.vocab.get_vocab_size("labels")
self.encoder = encoder
self.feedforward_layer = TimeDistributed(feedforward) if feedforward else None
self.pos_tag_embedding = pos_tag_embedding or None
if feedforward is not None:
output_dim = feedforward.get_output_dim()
else:
output_dim = span_extractor.get_output_dim()
self.tag_projection_layer = TimeDistributed(Linear(output_dim, self.num_classes))
representation_dim = text_field_embedder.get_output_dim()
if pos_tag_embedding is not None:
representation_dim += pos_tag_embedding.get_output_dim()
check_dimensions_match(
representation_dim,
encoder.get_input_dim(),
"representation dim (tokens + optional POS tags)",
"encoder input dim",
)
check_dimensions_match(
encoder.get_output_dim(),
span_extractor.get_input_dim(),
"encoder input dim",
"span extractor input dim",
)
if feedforward is not None:
check_dimensions_match(
span_extractor.get_output_dim(),
feedforward.get_input_dim(),
"span extractor output dim",
"feedforward input dim",
)
self.tag_accuracy = CategoricalAccuracy()
if evalb_directory_path is not None:
self._evalb_score = EvalbBracketingScorer(evalb_directory_path)
else:
self._evalb_score = None
initializer(self)
@overrides
def forward(
self, # type: ignore
tokens: TextFieldTensors,
spans: torch.LongTensor,
metadata: List[Dict[str, Any]],
pos_tags: TextFieldTensors = None,
span_labels: torch.LongTensor = None,
) -> Dict[str, torch.Tensor]:
"""
# Parameters
tokens : `TextFieldTensors`, required
The output of `TextField.as_array()`, which should typically be passed directly to a
`TextFieldEmbedder`. This output is a dictionary mapping keys to `TokenIndexer`
tensors. At its most basic, using a `SingleIdTokenIndexer` this is : `{"tokens":
Tensor(batch_size, num_tokens)}`. This dictionary will have the same keys as were used
for the `TokenIndexers` when you created the `TextField` representing your
sequence. The dictionary is designed to be passed directly to a `TextFieldEmbedder`,
which knows how to combine different word representations into a single vector per
token in your input.
spans : `torch.LongTensor`, required.
A tensor of shape `(batch_size, num_spans, 2)` representing the
inclusive start and end indices of all possible spans in the sentence.
metadata : `List[Dict[str, Any]]`, required.
A dictionary of metadata for each batch element which has keys:
tokens : `List[str]`, required.
The original string tokens in the sentence.
gold_tree : `nltk.Tree`, optional (default = `None`)
Gold NLTK trees for use in evaluation.
pos_tags : `List[str]`, optional.
The POS tags for the sentence. These can be used in the
model as embedded features, but they are passed here
in addition for use in constructing the tree.
pos_tags : `torch.LongTensor`, optional (default = `None`)
The output of a `SequenceLabelField` containing POS tags.
span_labels : `torch.LongTensor`, optional (default = `None`)
A torch tensor representing the integer gold class labels for all possible
spans, of shape `(batch_size, num_spans)`.
# Returns
An output dictionary consisting of:
class_probabilities : `torch.FloatTensor`
A tensor of shape `(batch_size, num_spans, span_label_vocab_size)`
representing a distribution over the label classes per span.
spans : `torch.LongTensor`
The original spans tensor.
tokens : `List[List[str]]`, required.
A list of tokens in the sentence for each element in the batch.
pos_tags : `List[List[str]]`, required.
A list of POS tags in the sentence for each element in the batch.
num_spans : `torch.LongTensor`, required.
A tensor of shape (batch_size), representing the lengths of non-padded spans
in `enumerated_spans`.
loss : `torch.FloatTensor`, optional
A scalar loss to be optimised.
"""
embedded_text_input = self.text_field_embedder(tokens)
if pos_tags is not None and self.pos_tag_embedding is not None:
embedded_pos_tags = self.pos_tag_embedding(pos_tags)
embedded_text_input = torch.cat([embedded_text_input, embedded_pos_tags], -1)
elif self.pos_tag_embedding is not None:
raise ConfigurationError("Model uses a POS embedding, but no POS tags were passed.")
mask = get_text_field_mask(tokens)
# Looking at the span start index is enough to know if
# this is padding or not. Shape: (batch_size, num_spans)
span_mask = (spans[:, :, 0] >= 0).squeeze(-1)
if span_mask.dim() == 1:
# This happens if you use batch_size 1 and encounter
# a length 1 sentence in PTB, which do exist. -.-
span_mask = span_mask.unsqueeze(-1)
if span_labels is not None and span_labels.dim() == 1:
span_labels = span_labels.unsqueeze(-1)
num_spans = get_lengths_from_binary_sequence_mask(span_mask)
encoded_text = self.encoder(embedded_text_input, mask)
span_representations = self.span_extractor(encoded_text, spans, mask, span_mask)
if self.feedforward_layer is not None:
span_representations = self.feedforward_layer(span_representations)
logits = self.tag_projection_layer(span_representations)
class_probabilities = masked_softmax(logits, span_mask.unsqueeze(-1))
output_dict = {
"class_probabilities": class_probabilities,
"spans": spans,
"tokens": [meta["tokens"] for meta in metadata],
"pos_tags": [meta.get("pos_tags") for meta in metadata],
"num_spans": num_spans,
}
if span_labels is not None:
loss = sequence_cross_entropy_with_logits(logits, span_labels, span_mask)
self.tag_accuracy(class_probabilities, span_labels, span_mask)
output_dict["loss"] = loss
# The evalb score is expensive to compute, so we only compute
# it for the validation and test sets.
batch_gold_trees = [meta.get("gold_tree") for meta in metadata]
if all(batch_gold_trees) and self._evalb_score is not None and not self.training:
gold_pos_tags: List[List[str]] = [
list(zip(*tree.pos()))[1] for tree in batch_gold_trees
]
predicted_trees = self.construct_trees(
class_probabilities.cpu().data,
spans.cpu().data,
num_spans.data,
output_dict["tokens"],
gold_pos_tags,
)
self._evalb_score(predicted_trees, batch_gold_trees)
return output_dict
@overrides
def make_output_human_readable(
self, output_dict: Dict[str, torch.Tensor]
) -> Dict[str, torch.Tensor]:
"""
Constructs an NLTK `Tree` given the scored spans. We also switch to exclusive
span ends when constructing the tree representation, because it makes indexing
into lists cleaner for ranges of text, rather than individual indices.
Finally, for batch prediction, we will have padded spans and class probabilities.
In order to make this less confusing, we remove all the padded spans and
distributions from `spans` and `class_probabilities` respectively.
"""
all_predictions = output_dict["class_probabilities"].cpu().data
all_spans = output_dict["spans"].cpu().data
all_sentences = output_dict["tokens"]
all_pos_tags = output_dict["pos_tags"] if all(output_dict["pos_tags"]) else None
num_spans = output_dict["num_spans"].data
trees = self.construct_trees(
all_predictions, all_spans, num_spans, all_sentences, all_pos_tags
)
batch_size = all_predictions.size(0)
output_dict["spans"] = [all_spans[i, : num_spans[i]] for i in range(batch_size)]
output_dict["class_probabilities"] = [
all_predictions[i, : num_spans[i], :] for i in range(batch_size)
]
output_dict["trees"] = trees
return output_dict
def construct_trees(
self,
predictions: torch.FloatTensor,
all_spans: torch.LongTensor,
num_spans: torch.LongTensor,
sentences: List[List[str]],
pos_tags: List[List[str]] = None,
) -> List[Tree]:
"""
Construct `nltk.Tree`'s for each batch element by greedily nesting spans.
The trees use exclusive end indices, which contrasts with how spans are
represented in the rest of the model.
# Parameters
predictions : `torch.FloatTensor`, required.
A tensor of shape `(batch_size, num_spans, span_label_vocab_size)`
representing a distribution over the label classes per span.
all_spans : `torch.LongTensor`, required.
A tensor of shape (batch_size, num_spans, 2), representing the span
indices we scored.
num_spans : `torch.LongTensor`, required.
A tensor of shape (batch_size), representing the lengths of non-padded spans
in `enumerated_spans`.
sentences : `List[List[str]]`, required.
A list of tokens in the sentence for each element in the batch.
pos_tags : `List[List[str]]`, optional (default = `None`).
A list of POS tags for each word in the sentence for each element
in the batch.
# Returns
A `List[Tree]` containing the decoded trees for each element in the batch.
"""
# Switch to using exclusive end spans.
exclusive_end_spans = all_spans.clone()
exclusive_end_spans[:, :, -1] += 1
no_label_id = self.vocab.get_token_index("NO-LABEL", "labels")
trees: List[Tree] = []
for batch_index, (scored_spans, spans, sentence) in enumerate(
zip(predictions, exclusive_end_spans, sentences)
):
selected_spans = []
for prediction, span in zip(
scored_spans[: num_spans[batch_index]], spans[: num_spans[batch_index]]
):
start, end = span
no_label_prob = prediction[no_label_id]
label_prob, label_index = torch.max(prediction, -1)
# Does the span have a label != NO-LABEL or is it the root node?
# If so, include it in the spans that we consider.
if int(label_index) != no_label_id or (start == 0 and end == len(sentence)):
selected_spans.append(
SpanInformation(
start=int(start),
end=int(end),
label_prob=float(label_prob),
no_label_prob=float(no_label_prob),
label_index=int(label_index),
)
)
# The spans we've selected might overlap, which causes problems when we try
# to construct the tree as they won't nest properly.
consistent_spans = self.resolve_overlap_conflicts_greedily(selected_spans)
spans_to_labels = {
(span.start, span.end): self.vocab.get_token_from_index(span.label_index, "labels")
for span in consistent_spans
}
sentence_pos = pos_tags[batch_index] if pos_tags is not None else None
trees.append(self.construct_tree_from_spans(spans_to_labels, sentence, sentence_pos))
return trees
@staticmethod
def resolve_overlap_conflicts_greedily(spans: List[SpanInformation]) -> List[SpanInformation]:
"""
Given a set of spans, removes spans which overlap by evaluating the difference
in probability between one being labeled and the other explicitly having no label
and vice-versa. The worst case time complexity of this method is `O(k * n^4)` where `n`
is the length of the sentence that the spans were enumerated from (and therefore
`k * m^2` complexity with respect to the number of spans `m`) and `k` is the
number of conflicts. However, in practice, there are very few conflicts. Hopefully.
This function modifies `spans` to remove overlapping spans.
# Parameters
spans : `List[SpanInformation]`, required.
A list of spans, where each span is a `namedtuple` containing the
following attributes:
start : `int`
The start index of the span.
end : `int`
The exclusive end index of the span.
no_label_prob : `float`
The probability of this span being assigned the `NO-LABEL` label.
label_prob : `float`
The probability of the most likely label.
# Returns
A modified list of `spans`, with the conflicts resolved by considering local
differences between pairs of spans and removing one of the two spans.
"""
conflicts_exist = True
while conflicts_exist:
conflicts_exist = False
for span1_index, span1 in enumerate(spans):
for span2_index, span2 in list(enumerate(spans))[span1_index + 1 :]:
if (
span1.start < span2.start < span1.end < span2.end
or span2.start < span1.start < span2.end < span1.end
):
# The spans overlap.
conflicts_exist = True
# What's the more likely situation: that span2 was labeled
# and span1 was unlabled, or that span1 was labeled and span2
# was unlabled? In the first case, we delete span2 from the
# set of spans to form the tree - in the second case, we delete
# span1.
if (
span1.no_label_prob + span2.label_prob
< span2.no_label_prob + span1.label_prob
):
spans.pop(span2_index)
else:
spans.pop(span1_index)
break
return spans
@staticmethod
def construct_tree_from_spans(
spans_to_labels: Dict[Tuple[int, int], str], sentence: List[str], pos_tags: List[str] = None
) -> Tree:
"""
# Parameters
spans_to_labels : `Dict[Tuple[int, int], str]`, required.
A mapping from spans to constituency labels.
sentence : `List[str]`, required.
A list of tokens forming the sentence to be parsed.
pos_tags : `List[str]`, optional (default = `None`)
A list of the pos tags for the words in the sentence, if they
were either predicted or taken as input to the model.
# Returns
An `nltk.Tree` constructed from the labelled spans.
"""
def assemble_subtree(start: int, end: int):
if (start, end) in spans_to_labels:
# Some labels contain nested spans, e.g S-VP.
# We actually want to create (S (VP ...)) nodes
# for these labels, so we split them up here.
labels: List[str] = spans_to_labels[(start, end)].split("-")
else:
labels = None
# This node is a leaf.
if end - start == 1:
word = sentence[start]
pos_tag = pos_tags[start] if pos_tags is not None else "XX"
tree = Tree(pos_tag, [word])
if labels is not None and pos_tags is not None:
# If POS tags were passed explicitly,
# they are added as pre-terminal nodes.
while labels:
tree = Tree(labels.pop(), [tree])
elif labels is not None:
# Otherwise, we didn't want POS tags
# at all.
tree = Tree(labels.pop(), [word])
while labels:
tree = Tree(labels.pop(), [tree])
return [tree]
argmax_split = start + 1
# Find the next largest subspan such that
# the left hand side is a constituent.
for split in range(end - 1, start, -1):
if (start, split) in spans_to_labels:
argmax_split = split
break
left_trees = assemble_subtree(start, argmax_split)
right_trees = assemble_subtree(argmax_split, end)
children = left_trees + right_trees
if labels is not None:
while labels:
children = [Tree(labels.pop(), children)]
return children
tree = assemble_subtree(0, len(sentence))
return tree[0]
@overrides
def get_metrics(self, reset: bool = False) -> Dict[str, float]:
all_metrics = {}
all_metrics["tag_accuracy"] = self.tag_accuracy.get_metric(reset=reset)
if self._evalb_score is not None:
evalb_metrics = self._evalb_score.get_metric(reset=reset)
all_metrics.update(evalb_metrics)
return all_metrics
default_predictor = "constituency_parser"
class SpanConstituencyParser(Model):
"""
This ``SpanConstituencyParser`` simply encodes a sequence of text
with a stacked ``Seq2SeqEncoder``, extracts span representations using a
``SpanExtractor``, and then predicts a label for each span in the sequence.
These labels are non-terminal nodes in a constituency parse tree, which we then
greedily reconstruct.
Parameters
----------
vocab : ``Vocabulary``, required
A Vocabulary, required in order to compute sizes for input/output projections.
text_field_embedder : ``TextFieldEmbedder``, required
Used to embed the ``tokens`` ``TextField`` we get as input to the model.
span_extractor : ``SpanExtractor``, required.
The method used to extract the spans from the encoded sequence.
encoder : ``Seq2SeqEncoder``, required.
The encoder that we will use in between embedding tokens and
generating span representations.
feedforward_layer : ``FeedForward``, required.
The FeedForward layer that we will use in between the encoder and the linear
projection to a distribution over span labels.
initializer : ``InitializerApplicator``, optional (default=``InitializerApplicator()``)
Used to initialize the model parameters.
regularizer : ``RegularizerApplicator``, optional (default=``None``)
If provided, will be used to calculate the regularization penalty during training.
"""
def __init__(self,
vocab: Vocabulary,
text_field_embedder: TextFieldEmbedder,
span_extractor: SpanExtractor,
encoder: Seq2SeqEncoder,
feedforward_layer: FeedForward = None,
initializer: InitializerApplicator = InitializerApplicator(),
regularizer: Optional[RegularizerApplicator] = None,
evalb_directory_path: str = None) -> None:
super(SpanConstituencyParser, self).__init__(vocab, regularizer)
self.text_field_embedder = text_field_embedder
self.span_extractor = span_extractor
self.num_classes = self.vocab.get_vocab_size("labels")
self.encoder = encoder
self.feedforward_layer = TimeDistributed(
feedforward_layer) if feedforward_layer else None
if feedforward_layer is not None:
output_dim = feedforward_layer.get_output_dim()
else:
output_dim = span_extractor.get_output_dim()
self.tag_projection_layer = TimeDistributed(
Linear(output_dim, self.num_classes))
check_dimensions_match(text_field_embedder.get_output_dim(),
encoder.get_input_dim(),
"text field embedding dim", "encoder input dim")
if feedforward_layer is not None:
check_dimensions_match(encoder.get_output_dim(),
feedforward_layer.get_input_dim(),
"stacked encoder output dim",
"feedforward input dim")
self.metrics = {
label: F1Measure(index)
for index, label in self.vocab.get_index_to_token_vocabulary(
"labels").items()
}
if evalb_directory_path is not None:
self._evalb_score = EvalbBracketingScorer(evalb_directory_path)
else:
self._evalb_score = None
initializer(self)
@overrides
def forward(
self, # type: ignore
tokens: Dict[str, torch.LongTensor],
spans: torch.LongTensor,
metadata: List[Dict[str, Any]],
span_labels: torch.LongTensor = None) -> Dict[str, torch.Tensor]:
# pylint: disable=arguments-differ
"""
Parameters
----------
tokens : Dict[str, torch.LongTensor], required
The output of ``TextField.as_array()``, which should typically be passed directly to a
``TextFieldEmbedder``. This output is a dictionary mapping keys to ``TokenIndexer``
tensors. At its most basic, using a ``SingleIdTokenIndexer`` this is: ``{"tokens":
Tensor(batch_size, num_tokens)}``. This dictionary will have the same keys as were used
for the ``TokenIndexers`` when you created the ``TextField`` representing your
sequence. The dictionary is designed to be passed directly to a ``TextFieldEmbedder``,
which knows how to combine different word representations into a single vector per
token in your input.
spans : ``torch.LongTensor``, required.
A tensor of shape ``(batch_size, num_spans, 2)`` representing the
inclusive start and end indices of all possible spans in the sentence.
span_labels : ``torch.LongTensor``, optional (default = None)
A torch tensor representing the integer gold class labels for all possible
spans, of shape ``(batch_size, num_spans)``.
metadata : List[Dict[str, Any]], required.
A dictionary of metadata for each batch element which has keys:
tokens : ``List[str]``, required.
The original string tokens in the sentence.
gold_tree : ``nltk.Tree``, optional (default = None)
Gold NLTK trees for use in evaluation.
Returns
-------
An output dictionary consisting of:
class_probabilities : ``torch.FloatTensor``
A tensor of shape ``(batch_size, num_spans, span_label_vocab_size)``
representing a distribution over the label classes per span.
spans : ``torch.LongTensor``
The original spans tensor.
tokens : ``List[List[str]]``, required.
A list of tokens in the sentence for each element in the batch.
num_spans : ``torch.LongTensor``, required.
A tensor of shape (batch_size), representing the lengths of non-padded spans
in ``enumerated_spans``.
loss : ``torch.FloatTensor``, optional
A scalar loss to be optimised.
"""
embedded_text_input = self.text_field_embedder(tokens)
mask = get_text_field_mask(tokens)
# Looking at the span start index is enough to know if
# this is padding or not. Shape: (batch_size, num_spans)
span_mask = (spans[:, :, 0] >= 0).squeeze(-1).long()
if span_mask.dim() == 1:
# This happens if you use batch_size 1 and encounter
# a length 1 sentence in PTB, which do exist. -.-
span_mask = span_mask.unsqueeze(-1)
num_spans = get_lengths_from_binary_sequence_mask(span_mask)
encoded_text = self.encoder(embedded_text_input, mask)
span_representations = self.span_extractor(encoded_text, spans, mask,
span_mask)
if self.feedforward_layer is not None:
span_representations = self.feedforward_layer(span_representations)
logits = self.tag_projection_layer(span_representations)
class_probabilities = last_dim_softmax(logits, span_mask.unsqueeze(-1))
output_dict = {
"class_probabilities": class_probabilities,
"spans": spans,
"tokens": [meta["tokens"] for meta in metadata],
"num_spans": num_spans
}
if span_labels is not None:
loss = sequence_cross_entropy_with_logits(logits, span_labels,
span_mask)
for metric in self.metrics.values():
metric(logits, span_labels, span_mask)
output_dict["loss"] = loss
# The evalb score is expensive to compute, so we only compute
# it for the validation and test sets.
batch_gold_trees = [meta.get("gold_tree") for meta in metadata]
if all(batch_gold_trees
) and self._evalb_score is not None and not self.training:
# TODO(Mark): Predict POS and use here instead of using the gold ones.
gold_pos_tags: List[List[str]] = [
list(zip(*tree.pos()))[1] for tree in batch_gold_trees
]
predicted_trees = self.construct_trees(
class_probabilities.cpu().data,
spans.cpu().data, num_spans.data, output_dict["tokens"],
gold_pos_tags)
self._evalb_score(predicted_trees, batch_gold_trees)
return output_dict
@overrides
def decode(
self, output_dict: Dict[str,
torch.Tensor]) -> Dict[str, torch.Tensor]:
"""
Constructs an NLTK ``Tree`` given the scored spans. We also switch to exclusive
span ends when constructing the tree representation, because it makes indexing
into lists cleaner for ranges of text, rather than individual indices.
Finally, for batch prediction, we will have padded spans and class probabilities.
In order to make this less confusing, we remove all the padded spans and
distributions from ``spans`` and ``class_probabilities`` respectively.
"""
all_predictions = output_dict['class_probabilities'].cpu().data
all_spans = output_dict["spans"].cpu().data
all_sentences = output_dict["tokens"]
num_spans = output_dict["num_spans"].data
trees = self.construct_trees(all_predictions, all_spans, num_spans,
all_sentences)
batch_size = all_predictions.size(0)
output_dict["spans"] = [
all_spans[i, :num_spans[i]] for i in range(batch_size)
]
output_dict["class_probabilities"] = [
all_predictions[i, :num_spans[i], :] for i in range(batch_size)
]
output_dict["trees"] = trees
return output_dict
def construct_trees(self,
predictions: torch.FloatTensor,
all_spans: torch.LongTensor,
num_spans: torch.LongTensor,
sentences: List[List[str]],
pos_tags: List[List[str]] = None) -> List[Tree]:
"""
Construct ``nltk.Tree``'s for each batch element by greedily nesting spans.
The trees use exclusive end indices, which contrasts with how spans are
represented in the rest of the model.
Parameters
----------
predictions : ``torch.FloatTensor``, required.
A tensor of shape ``(batch_size, num_spans, span_label_vocab_size)``
representing a distribution over the label classes per span.
all_spans : ``torch.LongTensor``, required.
A tensor of shape (batch_size, num_spans, 2), representing the span
indices we scored.
num_spans : ``torch.LongTensor``, required.
A tensor of shape (batch_size), representing the lengths of non-padded spans
in ``enumerated_spans``.
sentences : ``List[List[str]]``, required.
A list of tokens in the sentence for each element in the batch.
pos_tags : ``List[List[str]]``, optional (default = None).
A list of POS tags for each word in the sentence for each element
in the batch.
Returns
-------
A ``List[Tree]`` containing the decoded trees for each element in the batch.
"""
# Switch to using exclusive end spans.
exclusive_end_spans = all_spans.clone()
exclusive_end_spans[:, :, -1] += 1
no_label_id = self.vocab.get_token_index("NO-LABEL", "labels")
trees: List[Tree] = []
for batch_index, (scored_spans, spans, sentence) in enumerate(
zip(predictions, exclusive_end_spans, sentences)):
selected_spans = []
for prediction, span in zip(scored_spans[:num_spans[batch_index]],
spans[:num_spans[batch_index]]):
start, end = span
no_label_prob = prediction[no_label_id]
label_prob, label_index = torch.max(prediction, -1)
# Does the span have a label != NO-LABEL or is it the root node?
# If so, include it in the spans that we consider.
if int(label_index) != no_label_id or (start == 0 and end
== len(sentence)):
# TODO(Mark): Remove this once pylint sorts out named tuples.
# https://github.com/PyCQA/pylint/issues/1418
selected_spans.append(
SpanInformation(
start=int(start), # pylint: disable=no-value-for-parameter
end=int(end),
label_prob=float(label_prob),
no_label_prob=float(no_label_prob),
label_index=int(label_index)))
# The spans we've selected might overlap, which causes problems when we try
# to construct the tree as they won't nest properly.
consistent_spans = self.resolve_overlap_conflicts_greedily(
selected_spans)
spans_to_labels = {
(span.start, span.end):
self.vocab.get_token_from_index(span.label_index, "labels")
for span in consistent_spans
}
sentence_pos = pos_tags[
batch_index] if pos_tags is not None else None
trees.append(
self.construct_tree_from_spans(spans_to_labels, sentence,
sentence_pos))
return trees
@staticmethod
def resolve_overlap_conflicts_greedily(
spans: List[SpanInformation]) -> List[SpanInformation]:
"""
Given a set of spans, removes spans which overlap by evaluating the difference
in probability between one being labeled and the other explicitly having no label
and vice-versa. The worst case time complexity of this method is ``O(k * n^4)`` where ``n``
is the length of the sentence that the spans were enumerated from (and therefore
``k * m^2`` complexity with respect to the number of spans ``m``) and ``k`` is the
number of conflicts. However, in practice, there are very few conflicts. Hopefully.
This function modifies ``spans`` to remove overlapping spans.
Parameters
----------
spans: ``List[SpanInformation]``, required.
A list of spans, where each span is a ``namedtuple`` containing the
following attributes:
start : ``int``
The start index of the span.
end : ``int``
The exclusive end index of the span.
no_label_prob : ``float``
The probability of this span being assigned the ``NO-LABEL`` label.
label_prob : ``float``
The probability of the most likely label.
Returns
-------
A modified list of ``spans``, with the conflicts resolved by considering local
differences between pairs of spans and removing one of the two spans.
"""
conflicts_exist = True
while conflicts_exist:
conflicts_exist = False
for span1_index, span1 in enumerate(spans):
for span2_index, span2 in list(enumerate(spans))[span1_index +
1:]:
if (span1.start < span2.start < span1.end < span2.end or
span2.start < span1.start < span2.end < span1.end):
# The spans overlap.
conflicts_exist = True
# What's the more likely situation: that span2 was labeled
# and span1 was unlabled, or that span1 was labeled and span2
# was unlabled? In the first case, we delete span2 from the
# set of spans to form the tree - in the second case, we delete
# span1.
if (span1.no_label_prob + span2.label_prob <
span2.no_label_prob + span1.label_prob):
spans.pop(span2_index)
else:
spans.pop(span1_index)
break
return spans
@staticmethod
def construct_tree_from_spans(spans_to_labels: Dict[Tuple[int, int], str],
sentence: List[str],
pos_tags: List[str] = None) -> Tree:
"""
Parameters
----------
spans_to_labels : ``Dict[Tuple[int, int], str]``, required.
A mapping from spans to constituency labels.
sentence : ``List[str]``, required.
A list of tokens forming the sentence to be parsed.
pos_tags : ``List[str]``, optional (default = None)
A list of the pos tags for the words in the sentence, if they
were either predicted or taken as input to the model.
Returns
-------
An ``nltk.Tree`` constructed from the labelled spans.
"""
def assemble_subtree(start: int, end: int):
if (start, end) in spans_to_labels:
# Some labels contain nested spans, e.g S-VP.
# We actually want to create (S (VP ...)) nodes
# for these labels, so we split them up here.
labels: List[str] = spans_to_labels[(start, end)].split("-")
else:
labels = None
# This node is a leaf.
if end - start == 1:
word = sentence[start]
pos_tag = pos_tags[start] if pos_tags is not None else "XX"
tree = Tree(pos_tag, [word])
if labels is not None and pos_tags is not None:
# If POS tags were passed explicitly,
# they are added as pre-terminal nodes.
while labels:
tree = Tree(labels.pop(), [tree])
elif labels is not None:
# Otherwise, we didn't want POS tags
# at all.
tree = Tree(labels.pop(), [word])
while labels:
tree = Tree(labels.pop(), [tree])
return [tree]
argmax_split = start + 1
# Find the next largest subspan such that
# the left hand side is a constituent.
for split in range(end - 1, start, -1):
if (start, split) in spans_to_labels:
argmax_split = split
break
left_trees = assemble_subtree(start, argmax_split)
right_trees = assemble_subtree(argmax_split, end)
children = left_trees + right_trees
if labels is not None:
while labels:
children = [Tree(labels.pop(), children)]
return children
tree = assemble_subtree(0, len(sentence))
return tree[0]
@overrides
def get_metrics(self, reset: bool = False) -> Dict[str, float]:
all_metrics = {}
total_f1 = 0.0
total_precision = 0.0
total_recall = 0.0
# We'll just capture the average f1, precision and recall
# because there are 27 constituent types, which makes
# for very verbose console output.
for metric in self.metrics.values():
f1, precision, recall = metric.get_metric(reset) # pylint: disable=invalid-name
total_f1 += f1
total_precision += precision
total_recall += recall
num_metrics = len(self.metrics)
all_metrics["average_f1"] = total_f1 / num_metrics
all_metrics["average_precision"] = total_precision / num_metrics
all_metrics["average_recall"] = total_recall / num_metrics
if self._evalb_score is not None:
evalb_metrics = self._evalb_score.get_metric(reset=reset)
all_metrics.update(evalb_metrics)
return all_metrics
@classmethod
def from_params(cls, vocab: Vocabulary,
params: Params) -> 'SpanConstituencyParser':
embedder_params = params.pop("text_field_embedder")
text_field_embedder = TextFieldEmbedder.from_params(
vocab, embedder_params)
span_extractor = SpanExtractor.from_params(
params.pop("span_extractor"))
encoder = Seq2SeqEncoder.from_params(params.pop("encoder"))
feed_forward_params = params.pop("feedforward", None)
if feed_forward_params is not None:
feedforward_layer = FeedForward.from_params(feed_forward_params)
else:
feedforward_layer = None
initializer = InitializerApplicator.from_params(
params.pop('initializer', []))
regularizer = RegularizerApplicator.from_params(
params.pop('regularizer', []))
evalb_directory_path = params.pop("evalb_directory_path", None)
params.assert_empty(cls.__name__)
return cls(vocab=vocab,
text_field_embedder=text_field_embedder,
span_extractor=span_extractor,
encoder=encoder,
feedforward_layer=feedforward_layer,
initializer=initializer,
regularizer=regularizer,
evalb_directory_path=evalb_directory_path)
def setUp(self):
EvalbBracketingScorer.compile_evalb()
super(SpanConstituencyParserTest, self).setUp()
self.set_up_model(self.FIXTURES_ROOT / "constituency_parser" / "constituency_parser.json",
self.FIXTURES_ROOT / "data" / "example_ptb.trees")
def setUp(self):
super().setUp()
EvalbBracketingScorer.compile_evalb()
class SpanConstituencyParser(Model):
"""
This ``SpanConstituencyParser`` simply encodes a sequence of text
with a stacked ``Seq2SeqEncoder``, extracts span representations using a
``SpanExtractor``, and then predicts a label for each span in the sequence.
These labels are non-terminal nodes in a constituency parse tree, which we then
greedily reconstruct.
Parameters
----------
vocab : ``Vocabulary``, required
A Vocabulary, required in order to compute sizes for input/output projections.
text_field_embedder : ``TextFieldEmbedder``, required
Used to embed the ``tokens`` ``TextField`` we get as input to the model.
span_extractor : ``SpanExtractor``, required.
The method used to extract the spans from the encoded sequence.
encoder : ``Seq2SeqEncoder``, required.
The encoder that we will use in between embedding tokens and
generating span representations.
feedforward_layer : ``FeedForward``, required.
The FeedForward layer that we will use in between the encoder and the linear
projection to a distribution over span labels.
pos_tag_embedding : ``Embedding``, optional.
Used to embed the ``pos_tags`` ``SequenceLabelField`` we get as input to the model.
initializer : ``InitializerApplicator``, optional (default=``InitializerApplicator()``)
Used to initialize the model parameters.
regularizer : ``RegularizerApplicator``, optional (default=``None``)
If provided, will be used to calculate the regularization penalty during training.
"""
def __init__(self,
vocab: Vocabulary,
text_field_embedder: TextFieldEmbedder,
span_extractor: SpanExtractor,
encoder: Seq2SeqEncoder,
feedforward_layer: FeedForward = None,
pos_tag_embedding: Embedding = None,
initializer: InitializerApplicator = InitializerApplicator(),
regularizer: Optional[RegularizerApplicator] = None,
evalb_directory_path: str = None) -> None:
super(SpanConstituencyParser, self).__init__(vocab, regularizer)
self.text_field_embedder = text_field_embedder
self.span_extractor = span_extractor
self.num_classes = self.vocab.get_vocab_size("labels")
self.encoder = encoder
self.feedforward_layer = TimeDistributed(feedforward_layer) if feedforward_layer else None
self.pos_tag_embedding = pos_tag_embedding or None
if feedforward_layer is not None:
output_dim = feedforward_layer.get_output_dim()
else:
output_dim = span_extractor.get_output_dim()
self.tag_projection_layer = TimeDistributed(Linear(output_dim, self.num_classes))
representation_dim = text_field_embedder.get_output_dim()
if pos_tag_embedding is not None:
representation_dim += pos_tag_embedding.get_output_dim()
check_dimensions_match(representation_dim,
encoder.get_input_dim(),
"representation dim (tokens + optional POS tags)",
"encoder input dim")
check_dimensions_match(encoder.get_output_dim(),
span_extractor.get_input_dim(),
"encoder input dim",
"span extractor input dim")
if feedforward_layer is not None:
check_dimensions_match(span_extractor.get_output_dim(),
feedforward_layer.get_input_dim(),
"span extractor output dim",
"feedforward input dim")
self.tag_accuracy = CategoricalAccuracy()
if evalb_directory_path is not None:
self._evalb_score = EvalbBracketingScorer(evalb_directory_path)
else:
self._evalb_score = None
initializer(self)
@overrides
def forward(self, # type: ignore
tokens: Dict[str, torch.LongTensor],
spans: torch.LongTensor,
metadata: List[Dict[str, Any]],
pos_tags: Dict[str, torch.LongTensor] = None,
span_labels: torch.LongTensor = None) -> Dict[str, torch.Tensor]:
# pylint: disable=arguments-differ
"""
Parameters
----------
tokens : Dict[str, torch.LongTensor], required
The output of ``TextField.as_array()``, which should typically be passed directly to a
``TextFieldEmbedder``. This output is a dictionary mapping keys to ``TokenIndexer``
tensors. At its most basic, using a ``SingleIdTokenIndexer`` this is: ``{"tokens":
Tensor(batch_size, num_tokens)}``. This dictionary will have the same keys as were used
for the ``TokenIndexers`` when you created the ``TextField`` representing your
sequence. The dictionary is designed to be passed directly to a ``TextFieldEmbedder``,
which knows how to combine different word representations into a single vector per
token in your input.
spans : ``torch.LongTensor``, required.
A tensor of shape ``(batch_size, num_spans, 2)`` representing the
inclusive start and end indices of all possible spans in the sentence.
metadata : List[Dict[str, Any]], required.
A dictionary of metadata for each batch element which has keys:
tokens : ``List[str]``, required.
The original string tokens in the sentence.
gold_tree : ``nltk.Tree``, optional (default = None)
Gold NLTK trees for use in evaluation.
pos_tags : ``List[str]``, optional.
The POS tags for the sentence. These can be used in the
model as embedded features, but they are passed here
in addition for use in constructing the tree.
pos_tags : ``torch.LongTensor``, optional (default = None)
The output of a ``SequenceLabelField`` containing POS tags.
span_labels : ``torch.LongTensor``, optional (default = None)
A torch tensor representing the integer gold class labels for all possible
spans, of shape ``(batch_size, num_spans)``.
Returns
-------
An output dictionary consisting of:
class_probabilities : ``torch.FloatTensor``
A tensor of shape ``(batch_size, num_spans, span_label_vocab_size)``
representing a distribution over the label classes per span.
spans : ``torch.LongTensor``
The original spans tensor.
tokens : ``List[List[str]]``, required.
A list of tokens in the sentence for each element in the batch.
pos_tags : ``List[List[str]]``, required.
A list of POS tags in the sentence for each element in the batch.
num_spans : ``torch.LongTensor``, required.
A tensor of shape (batch_size), representing the lengths of non-padded spans
in ``enumerated_spans``.
loss : ``torch.FloatTensor``, optional
A scalar loss to be optimised.
"""
embedded_text_input = self.text_field_embedder(tokens)
if pos_tags is not None and self.pos_tag_embedding is not None:
embedded_pos_tags = self.pos_tag_embedding(pos_tags)
embedded_text_input = torch.cat([embedded_text_input, embedded_pos_tags], -1)
elif self.pos_tag_embedding is not None:
raise ConfigurationError("Model uses a POS embedding, but no POS tags were passed.")
mask = get_text_field_mask(tokens)
# Looking at the span start index is enough to know if
# this is padding or not. Shape: (batch_size, num_spans)
span_mask = (spans[:, :, 0] >= 0).squeeze(-1).long()
if span_mask.dim() == 1:
# This happens if you use batch_size 1 and encounter
# a length 1 sentence in PTB, which do exist. -.-
span_mask = span_mask.unsqueeze(-1)
if span_labels is not None and span_labels.dim() == 1:
span_labels = span_labels.unsqueeze(-1)
num_spans = get_lengths_from_binary_sequence_mask(span_mask)
encoded_text = self.encoder(embedded_text_input, mask)
span_representations = self.span_extractor(encoded_text, spans, mask, span_mask)
if self.feedforward_layer is not None:
span_representations = self.feedforward_layer(span_representations)
logits = self.tag_projection_layer(span_representations)
class_probabilities = last_dim_softmax(logits, span_mask.unsqueeze(-1))
output_dict = {
"class_probabilities": class_probabilities,
"spans": spans,
"tokens": [meta["tokens"] for meta in metadata],
"pos_tags": [meta.get("pos_tags") for meta in metadata],
"num_spans": num_spans
}
if span_labels is not None:
loss = sequence_cross_entropy_with_logits(logits, span_labels, span_mask)
self.tag_accuracy(class_probabilities, span_labels, span_mask)
output_dict["loss"] = loss
# The evalb score is expensive to compute, so we only compute
# it for the validation and test sets.
batch_gold_trees = [meta.get("gold_tree") for meta in metadata]
if all(batch_gold_trees) and self._evalb_score is not None and not self.training:
gold_pos_tags: List[List[str]] = [list(zip(*tree.pos()))[1]
for tree in batch_gold_trees]
predicted_trees = self.construct_trees(class_probabilities.cpu().data,
spans.cpu().data,
num_spans.data,
output_dict["tokens"],
gold_pos_tags)
self._evalb_score(predicted_trees, batch_gold_trees)
return output_dict
@overrides
def decode(self, output_dict: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]:
"""
Constructs an NLTK ``Tree`` given the scored spans. We also switch to exclusive
span ends when constructing the tree representation, because it makes indexing
into lists cleaner for ranges of text, rather than individual indices.
Finally, for batch prediction, we will have padded spans and class probabilities.
In order to make this less confusing, we remove all the padded spans and
distributions from ``spans`` and ``class_probabilities`` respectively.
"""
all_predictions = output_dict['class_probabilities'].cpu().data
all_spans = output_dict["spans"].cpu().data
all_sentences = output_dict["tokens"]
all_pos_tags = output_dict["pos_tags"] if all(output_dict["pos_tags"]) else None
num_spans = output_dict["num_spans"].data
trees = self.construct_trees(all_predictions, all_spans, num_spans, all_sentences, all_pos_tags)
batch_size = all_predictions.size(0)
output_dict["spans"] = [all_spans[i, :num_spans[i]] for i in range(batch_size)]
output_dict["class_probabilities"] = [all_predictions[i, :num_spans[i], :] for i in range(batch_size)]
output_dict["trees"] = trees
return output_dict
def construct_trees(self,
predictions: torch.FloatTensor,
all_spans: torch.LongTensor,
num_spans: torch.LongTensor,
sentences: List[List[str]],
pos_tags: List[List[str]] = None) -> List[Tree]:
"""
Construct ``nltk.Tree``'s for each batch element by greedily nesting spans.
The trees use exclusive end indices, which contrasts with how spans are
represented in the rest of the model.
Parameters
----------
predictions : ``torch.FloatTensor``, required.
A tensor of shape ``(batch_size, num_spans, span_label_vocab_size)``
representing a distribution over the label classes per span.
all_spans : ``torch.LongTensor``, required.
A tensor of shape (batch_size, num_spans, 2), representing the span
indices we scored.
num_spans : ``torch.LongTensor``, required.
A tensor of shape (batch_size), representing the lengths of non-padded spans
in ``enumerated_spans``.
sentences : ``List[List[str]]``, required.
A list of tokens in the sentence for each element in the batch.
pos_tags : ``List[List[str]]``, optional (default = None).
A list of POS tags for each word in the sentence for each element
in the batch.
Returns
-------
A ``List[Tree]`` containing the decoded trees for each element in the batch.
"""
# Switch to using exclusive end spans.
exclusive_end_spans = all_spans.clone()
exclusive_end_spans[:, :, -1] += 1
no_label_id = self.vocab.get_token_index("NO-LABEL", "labels")
trees: List[Tree] = []
for batch_index, (scored_spans, spans, sentence) in enumerate(zip(predictions,
exclusive_end_spans,
sentences)):
selected_spans = []
for prediction, span in zip(scored_spans[:num_spans[batch_index]],
spans[:num_spans[batch_index]]):
start, end = span
no_label_prob = prediction[no_label_id]
label_prob, label_index = torch.max(prediction, -1)
# Does the span have a label != NO-LABEL or is it the root node?
# If so, include it in the spans that we consider.
if int(label_index) != no_label_id or (start == 0 and end == len(sentence)):
# TODO(Mark): Remove this once pylint sorts out named tuples.
# https://github.com/PyCQA/pylint/issues/1418
selected_spans.append(SpanInformation(start=int(start), # pylint: disable=no-value-for-parameter
end=int(end),
label_prob=float(label_prob),
no_label_prob=float(no_label_prob),
label_index=int(label_index)))
# The spans we've selected might overlap, which causes problems when we try
# to construct the tree as they won't nest properly.
consistent_spans = self.resolve_overlap_conflicts_greedily(selected_spans)
spans_to_labels = {(span.start, span.end):
self.vocab.get_token_from_index(span.label_index, "labels")
for span in consistent_spans}
sentence_pos = pos_tags[batch_index] if pos_tags is not None else None
trees.append(self.construct_tree_from_spans(spans_to_labels, sentence, sentence_pos))
return trees
@staticmethod
def resolve_overlap_conflicts_greedily(spans: List[SpanInformation]) -> List[SpanInformation]:
"""
Given a set of spans, removes spans which overlap by evaluating the difference
in probability between one being labeled and the other explicitly having no label
and vice-versa. The worst case time complexity of this method is ``O(k * n^4)`` where ``n``
is the length of the sentence that the spans were enumerated from (and therefore
``k * m^2`` complexity with respect to the number of spans ``m``) and ``k`` is the
number of conflicts. However, in practice, there are very few conflicts. Hopefully.
This function modifies ``spans`` to remove overlapping spans.
Parameters
----------
spans: ``List[SpanInformation]``, required.
A list of spans, where each span is a ``namedtuple`` containing the
following attributes:
start : ``int``
The start index of the span.
end : ``int``
The exclusive end index of the span.
no_label_prob : ``float``
The probability of this span being assigned the ``NO-LABEL`` label.
label_prob : ``float``
The probability of the most likely label.
Returns
-------
A modified list of ``spans``, with the conflicts resolved by considering local
differences between pairs of spans and removing one of the two spans.
"""
conflicts_exist = True
while conflicts_exist:
conflicts_exist = False
for span1_index, span1 in enumerate(spans):
for span2_index, span2 in list(enumerate(spans))[span1_index + 1:]:
if (span1.start < span2.start < span1.end < span2.end or
span2.start < span1.start < span2.end < span1.end):
# The spans overlap.
conflicts_exist = True
# What's the more likely situation: that span2 was labeled
# and span1 was unlabled, or that span1 was labeled and span2
# was unlabled? In the first case, we delete span2 from the
# set of spans to form the tree - in the second case, we delete
# span1.
if (span1.no_label_prob + span2.label_prob <
span2.no_label_prob + span1.label_prob):
spans.pop(span2_index)
else:
spans.pop(span1_index)
break
return spans
@staticmethod
def construct_tree_from_spans(spans_to_labels: Dict[Tuple[int, int], str],
sentence: List[str],
pos_tags: List[str] = None) -> Tree:
"""
Parameters
----------
spans_to_labels : ``Dict[Tuple[int, int], str]``, required.
A mapping from spans to constituency labels.
sentence : ``List[str]``, required.
A list of tokens forming the sentence to be parsed.
pos_tags : ``List[str]``, optional (default = None)
A list of the pos tags for the words in the sentence, if they
were either predicted or taken as input to the model.
Returns
-------
An ``nltk.Tree`` constructed from the labelled spans.
"""
def assemble_subtree(start: int, end: int):
if (start, end) in spans_to_labels:
# Some labels contain nested spans, e.g S-VP.
# We actually want to create (S (VP ...)) nodes
# for these labels, so we split them up here.
labels: List[str] = spans_to_labels[(start, end)].split("-")
else:
labels = None
# This node is a leaf.
if end - start == 1:
word = sentence[start]
pos_tag = pos_tags[start] if pos_tags is not None else "XX"
tree = Tree(pos_tag, [word])
if labels is not None and pos_tags is not None:
# If POS tags were passed explicitly,
# they are added as pre-terminal nodes.
while labels:
tree = Tree(labels.pop(), [tree])
elif labels is not None:
# Otherwise, we didn't want POS tags
# at all.
tree = Tree(labels.pop(), [word])
while labels:
tree = Tree(labels.pop(), [tree])
return [tree]
argmax_split = start + 1
# Find the next largest subspan such that
# the left hand side is a constituent.
for split in range(end - 1, start, -1):
if (start, split) in spans_to_labels:
argmax_split = split
break
left_trees = assemble_subtree(start, argmax_split)
right_trees = assemble_subtree(argmax_split, end)
children = left_trees + right_trees
if labels is not None:
while labels:
children = [Tree(labels.pop(), children)]
return children
tree = assemble_subtree(0, len(sentence))
return tree[0]
@overrides
def get_metrics(self, reset: bool = False) -> Dict[str, float]:
all_metrics = {}
all_metrics["tag_accuracy"] = self.tag_accuracy.get_metric(reset=reset)
if self._evalb_score is not None:
evalb_metrics = self._evalb_score.get_metric(reset=reset)
all_metrics.update(evalb_metrics)
return all_metrics
@classmethod
def from_params(cls, vocab: Vocabulary, params: Params) -> 'SpanConstituencyParser':
embedder_params = params.pop("text_field_embedder")
text_field_embedder = TextFieldEmbedder.from_params(vocab, embedder_params)
span_extractor = SpanExtractor.from_params(params.pop("span_extractor"))
encoder = Seq2SeqEncoder.from_params(params.pop("encoder"))
feed_forward_params = params.pop("feedforward", None)
if feed_forward_params is not None:
feedforward_layer = FeedForward.from_params(feed_forward_params)
else:
feedforward_layer = None
pos_tag_embedding_params = params.pop("pos_tag_embedding", None)
if pos_tag_embedding_params is not None:
pos_tag_embedding = Embedding.from_params(vocab, pos_tag_embedding_params)
else:
pos_tag_embedding = None
initializer = InitializerApplicator.from_params(params.pop('initializer', []))
regularizer = RegularizerApplicator.from_params(params.pop('regularizer', []))
evalb_directory_path = params.pop("evalb_directory_path", None)
params.assert_empty(cls.__name__)
return cls(vocab=vocab,
text_field_embedder=text_field_embedder,
span_extractor=span_extractor,
encoder=encoder,
feedforward_layer=feedforward_layer,
pos_tag_embedding=pos_tag_embedding,
initializer=initializer,
regularizer=regularizer,
evalb_directory_path=evalb_directory_path)
def tearDown(self):
EvalbBracketingScorer.clean_evalb()
super(SpanConstituencyParserTest, self).tearDown()
def tearDown(self):
EvalbBracketingScorer.clean_evalb()
super().tearDown()
def setUp(self):
super().setUp()
EvalbBracketingScorer.compile_evalb()
