class Parser:
def __init__(self, archive_path, cuda_device, overrides, weights_file,
lookup_path, wordnet_path, am_tools_path):
jnius_config.set_classpath(".", am_tools_path)
# Load model
archive = load_archive(archive_path, cuda_device, overrides,
weights_file)
config = archive.config
config.formalism = "DUMMY"
prepare_environment(config)
model = archive.model
model.eval()
dataset_reader = DatasetReader.from_params(
config.pop('dataset_reader'))
self.predictor = AMconllPredictor(dataset_reader,
k=6,
give_up=0,
threads=1,
model=model)
self.formalism = AMRInterface(lookup_path, wordnet_path)
def parse(self, sentence):
print("")
words = spacy_tokenize(sentence)
am_sentence = from_raw_text(rawstr=sentence,
words=words,
add_art_root=False,
attributes=dict(),
contract_ne=True)
with TemporaryDirectory() as direc:
temp_path = direc + f"/sentences.amconll"
output_filename = direc + "/parsed.amconll"
with open(temp_path, "w") as f:
f.write(str(am_sentence))
f.write("\n\n")
self.predictor.parse_and_save(formalism="AMR-2017",
input_file=temp_path,
output_file=output_filename)
interpreted_output = self.formalism.evaluate(
output_filename, direc)
with open(interpreted_output) as f:
return str(f.read()).rstrip()
def __init__(self, archive_path, cuda_device, overrides, weights_file,
lookup_path, wordnet_path, am_tools_path):
jnius_config.set_classpath(".", am_tools_path)
# Load model
archive = load_archive(archive_path, cuda_device, overrides,
weights_file)
config = archive.config
config.formalism = "DUMMY"
prepare_environment(config)
model = archive.model
model.eval()
dataset_reader = DatasetReader.from_params(
config.pop('dataset_reader'))
self.predictor = AMconllPredictor(dataset_reader,
k=6,
give_up=0,
threads=1,
model=model)
self.formalism = AMRInterface(lookup_path, wordnet_path)
else:
dataset_reader = DatasetReader.from_params(config.pop('dataset_reader'))
evaluation_data_path = args.input_file
embedding_sources: Dict[str, str] = (json.loads(args.embedding_sources_mapping)
if args.embedding_sources_mapping else {})
if args.extend_vocab:
logger.info("Vocabulary is being extended with test instances.")
logger.info("Reading evaluation data from %s", evaluation_data_path)
instances = dataset_reader.read(evaluation_data_path)
model.vocab.extend_from_instances(Params({}), instances=instances)
model.extend_embedder_vocab(embedding_sources)
predictor = AMconllPredictor(dataset_reader,
args.k,
args.give_up,
args.threads,
model=model)
requires_art_root = {
"DM": True,
"PAS": True,
"PSD": True,
"EDS": False,
"AMR-2015": False,
"AMR-2017": False
}
sentences = []
with open(args.input_file) as f:
for sentence in f:
else:
dataset_reader = DatasetReader.from_params(config.pop('dataset_reader'))
evaluation_data_path = args.input_file
embedding_sources: Dict[str, str] = (json.loads(args.embedding_sources_mapping)
if args.embedding_sources_mapping else {})
if args.extend_vocab:
logger.info("Vocabulary is being extended with test instances.")
logger.info("Reading evaluation data from %s", evaluation_data_path)
instances = dataset_reader.read(evaluation_data_path)
model.vocab.extend_from_instances(Params({}), instances=instances)
model.extend_embedder_vocab(embedding_sources)
predictor = AMconllPredictor(dataset_reader,
args.k,
args.give_up,
args.threads,
model=model)
formalism = args.formalism
instances = dataset_reader.read([[
formalism, args.input_file
]]) # we need to give the formalism to amconll dataset_reader
model.train(False)
data_iterator = DataIterator.from_params(config.pop('iterator'))
with open(args.input_file) as f:
conll_sentences = list(amconll_tools.parse_amconll(f))
predictions = dataset_reader.restore_order(
forward_on_instances(model, instances, data_iterator))