def predict_model_with_archive(predictor: str,
params: Params,
archive: str,
input_file: str,
output_file: str,
batch_size: int = None):
if 'cuda_device' in params['trainer']:
cuda_device = params['trainer']['cuda_device']
from allennlp.common.checks import check_for_gpu
check_for_gpu(cuda_device)
archive = load_archive(archive, cuda_device=cuda_device)
else:
archive = load_archive(archive)
for item in archive.config.duplicate():
archive.config.__delitem__(item)
for item in params:
archive.config[item] = params.as_dict()[item]
predictor = MachampPredictor.from_archive(archive, predictor)
if batch_size == None:
batch_size = params['data_loader']['batch_sampler']['batch_size']
manager = _PredictManager(predictor,
input_file,
output_file,
batch_size,
print_to_console=False,
has_dataset_reader=True)
manager.run()
def predict_model(predictor: str,
params: Params,
archive_dir: str,
input_file: str,
output_file: str,
batch_size: int = 1):
"""
Predict output annotations from the given model and input file and produce an output file.
:param predictor: the type of predictor to use, e.g., "udify_predictor"
:param params: the Params of the model
:param archive_dir: the saved model archive
:param input_file: the input file to predict
:param output_file: the output file to save
:param batch_size: the batch size, set this higher to speed up GPU inference
"""
cuda_device = params["trainer"]["cuda_device"]
check_for_gpu(cuda_device)
archive = load_archive(os.path.join(archive_dir, "model.tar.gz"),
cuda_device=cuda_device)
predictor = Predictor.from_archive(archive, predictor)
manager = _PredictManager(predictor,
input_file,
output_file,
batch_size,
print_to_console=False,
has_dataset_reader=True)
manager.run()
def file_iface_predictor(input_file, output_file):
manager = _PredictManager(predictor,
input_file,
output_file,
batch_size,
print_to_console=False,
has_dataset_reader=True)
manager.run()
def predict(self, args: argparse.Namespace, predictor: Predictor):
manager = _PredictManager(
predictor,
args.input_file,
args.output_file,
args.batch_size,
not args.silent,
args.use_dataset_reader,
)
manager.run()
def predict_model_with_archive(predictor: str, params: Params, archive: str,
input_file: str, output_file: str, batch_size: int = 1):
cuda_device = params["trainer"]["cuda_device"]
check_for_gpu(cuda_device)
archive = load_archive(archive,
cuda_device=cuda_device)
predictor = Predictor.from_archive(archive, predictor)
manager = _PredictManager(predictor,
input_file,
output_file,
batch_size,
print_to_console=False,
has_dataset_reader=True)
manager.run()
def init_load_model():
"""
The [`run`](./train.md#run) command only knows about the registered classes in the ``allennlp``
codebase. In particular, once you start creating your own `Model` s and so forth, it won't
work for them, unless you use the ``--include-package`` flag or you make your code available
as a plugin (see [`plugins`](./plugins.md)).
"""
global predict_manager, cuda_place
import_plugins()
parser = create_parser("allennlp")
args = parser.parse_args()
args.archive_file = r"models/ace05_event/model.tar.gz"
args.cuda_device = cuda_place
args.input_file = r"data/ace-event/processed-data/default-settings/json/test.json"
args.output_file = r"predictions/event_predict.json"
args.use_dataset_reader = True
args.include_package = "dygie"
args.predictor = "dygie"
args.weights_file = None
args.overrides = ''
args.dataset_reader_choice = 'validation'
args.batch_size = 1
args.silent = False
# print(dir(args))
# print(args.include_package)
# for package_name in args.include_package:
# import_module_and_submodules_new(package_name)
import_module_and_submodules_new('dygie')
import_module_and_submodules_new('dygie.predictors.dygie')
import_module_and_submodules_new('dygie.models.dygie')
# predictor = DyGIEPredictor.from_path(os.path.join(current_path, 'models/ace05_event/model.tar.gz'), predictor_name="dygie")
predictor = predict._get_predictor(args)
predict_manager = predict._PredictManager(
predictor,
args.input_file,
args.output_file,
args.batch_size,
not args.silent,
args.use_dataset_reader,
)
def pred(cuda_device=0,
archive_file="/backup3/jcxu/exComp/tmp_expsc74o5pf7/model.tar.gz",
weights_file="/backup3/jcxu/exComp/tmp_expsc74o5pf7/best.th",
predictor='lstm-tagger',
input_file="/backup3/jcxu/exComp/example.txt"):
with open(input_file, 'w') as fd:
json.dump({"sentence": "This is a useful sentence."}, fd)
fd.write("\n")
json.dump({"sentence": "This is a gree, blue and useful sentence."},
fd)
fd.write("\n")
json.dump({"sentence": "This is a useless sentence."}, fd)
check_for_gpu(cuda_device)
archive = load_archive(archive_file,
weights_file=weights_file,
cuda_device=cuda_device,
overrides="")
# predictor = SentenceTaggerPredictor(archive, dataset_reader=PosDatasetReader())
predictor = Predictor.from_archive(archive, 'sentence-tagger')
manager = _PredictManager(predictor, input_file, None, 1, not False, False)
manager.run()
def run(_):
"""Run model."""
# Imports are required to make Registrable modules visible without passing parameter
util.import_module_and_submodules("combo.commands")
util.import_module_and_submodules("combo.models")
util.import_module_and_submodules("combo.training")
if FLAGS.mode == "train":
checks.file_exists(FLAGS.config_path)
params = common.Params.from_file(FLAGS.config_path,
ext_vars=_get_ext_vars())
model_params = params.get("model").as_ordered_dict()
serialization_dir = tempfile.mkdtemp(prefix="allennlp",
dir=FLAGS.serialization_dir)
model = train.train_model(params,
serialization_dir=serialization_dir,
file_friendly_logging=True)
logger.info(f"Training model stored in: {serialization_dir}")
if FLAGS.finetuning_training_data_path:
for f in FLAGS.finetuning_training_data_path:
checks.file_exists(f)
# Loading will be performed from stored model.tar.gz
del model
if torch.cuda.is_available():
torch.cuda.empty_cache()
params = common.Params.from_file(
FLAGS.config_path, ext_vars=_get_ext_vars(finetuning=True))
# Replace model definition with pretrained archive
params["model"] = {
"type": "from_archive",
"archive_file": serialization_dir + "/model.tar.gz",
}
serialization_dir = tempfile.mkdtemp(prefix="allennlp",
suffix="-finetuning",
dir=FLAGS.serialization_dir)
model = train.train_model(params.duplicate(),
serialization_dir=serialization_dir,
file_friendly_logging=True)
# Make finetuning model serialization independent from training serialization
# Storing model definition instead of archive
params["model"] = model_params
params.to_file(
os.path.join(serialization_dir, archival.CONFIG_NAME))
archival.archive_model(serialization_dir)
logger.info(f"Finetuned model stored in: {serialization_dir}")
if FLAGS.test_path and FLAGS.output_file:
checks.file_exists(FLAGS.test_path)
params = common.Params.from_file(
FLAGS.config_path, ext_vars=_get_ext_vars())["dataset_reader"]
params.pop("type")
dataset_reader = dataset.UniversalDependenciesDatasetReader.from_params(
params)
predictor = predict.SemanticMultitaskPredictor(
model=model, dataset_reader=dataset_reader)
test_trees = dataset_reader.read(FLAGS.test_path)
with open(FLAGS.output_file, "w") as file:
for tree in test_trees:
file.writelines(
api.sentence2conllu(
predictor.predict_instance(tree),
keep_semrel=dataset_reader.use_sem).serialize())
else:
use_dataset_reader = FLAGS.conllu_format
predictor = _get_predictor()
if FLAGS.input_file == "-":
use_dataset_reader = False
predictor.without_sentence_embedding = True
if use_dataset_reader:
predictor.line_to_conllu = True
if FLAGS.silent:
logging.getLogger("allennlp.common.params").disabled = True
manager = allen_predict._PredictManager(
predictor,
FLAGS.input_file,
FLAGS.output_file,
FLAGS.batch_size,
not FLAGS.silent,
use_dataset_reader,
)
manager.run()
