print()
string_labels = list(map(lambda x: label_to_cat[x], dataset.labels))
print(f"Labels distribution: {Counter(string_labels)}")
print()
print(f"Dataset size: {len(dataset)}")
print()
LABELS_TO_ID = dataset.label_to_id
print(f"Labels mapping: {LABELS_TO_ID}")
print()
metrics = {"validation": [AccuracyMeter]}
model_config = config.SenseModelParameters(model_name=args.config_name,
hidden_size=args.hidden_size,
num_classes=len(LABELS_TO_ID),
freeze_weights=False,
context_layers=(-1, ))
configuration = config.Configuration(
model_parameters=model_config,
model=args.model,
save_path=args.save_path,
sequence_max_len=args.seq_len,
batch_size=args.batch_size,
epochs=args.epochs,
device=torch.device(args.device),
tokenizer=transformers.AutoTokenizer.from_pretrained(args.model),
)
"""
model = TransformerWrapper.load_pretrained(
type=str,
dest="direction",
default="minimize")
#parser.add_argument('--n_splits', type=int, dest="n_splits", default=3, required=False)
#parser.add_argument('--fold', type=int, dest="fold", default=0, required=True)
args = parser.parse_args()
random.seed(43)
metrics = {"training": [AccuracyMeter], "validation": [AccuracyMeter]}
model_config = config.SenseModelParameters(
model_name=args.config_name,
hidden_size=args.hidden_size,
num_classes=len(CATEGORIES),
use_pretrained_embeddings=args.use_pretrained_embeddings,
freeze_weights=False,
context_layers=(-1, ))
configuration = config.Configuration(
model_parameters=model_config,
model=args.model,
save_path=args.save_path,
sequence_max_len=args.seq_len,
dropout_prob=args.dropout,
lr=args.lr,
batch_size=args.batch_size,
epochs=args.epochs,
device=torch.device(args.device),
tokenizer=transformers.AutoTokenizer.from_pretrained(args.model),
args = parser.parse_args()
processor = dataset.WicProcessor()
train_dataset = processor.build_dataset(args.train_path,
args.gold_train_path)
valid_dataset = processor.build_dataset(args.valid_path,
args.gold_valid_path)
train_data_loader = dataset.WiCDataLoader.build_batches(
train_dataset, args.batch_size)
valid_data_loader = dataset.WiCDataLoader.build_batches(
valid_dataset, args.batch_size)
model_config = config.SenseModelParameters(
model_name=args.config_name,
hidden_size=args.hidden_size,
num_classes=2,
use_pretrained_embeddings=args.use_pretrained_embeddings,
freeze_weights=args.freeze_weights,
context_layers=(-1, -2, -3, -4))
configuration = config.Configuration(
model_parameters=model_config,
model=args.model,
save_path=args.save_path,
sequence_max_len=args.seq_len,
dropout_prob=args.dropout,
lr=args.lr,
batch_size=args.batch_size,
epochs=args.epochs,
device=torch.device(args.device),
embedding_map=config.CONFIG.embedding_map,
default="standard")
parser.add_argument('--sense_features',
type=bool,
dest="senses_as_features",
default=False)
POOLING_STRATEGIES = {
"standard": SequencePoolingStrategy,
"sense": WordSensePoolingStrategy
}
args = parser.parse_args()
model_config = config.SenseModelParameters(
model_name=args.config_name,
hidden_size=args.hidden_size,
num_classes=len(LABELS_MAPPING),
use_pretrained_embeddings=args.use_pretrained_embeddings)
configuration = config.WordModelConfiguration(
model_parameters=model_config,
model=args.model,
save_path=args.save_path,
sequence_max_len=args.seq_len,
dropout_prob=args.dropout,
lr=args.lr,
batch_size=args.batch_size,
epochs=args.epochs,
device=torch.device(args.device),
embedding_map=None,
bnids_map=None,
#save_file(train_data_loader, "../dataset/cached/", "train_jp-pawsx-16-softmax")
#save_file(valid_data_loader, "../dataset/cached/", "valid_jp-pawsx-16-softmax")
metrics = ({
"training": [AccuracyMeter],
"validation": [AccuracyMeter]
} if args.loss == "softmax" else {
"training": [SimilarityAveragePrecisionMeter, SimilarityAccuracyMeter],
"validation":
[SimilarityAveragePrecisionMeter, SimilarityAccuracyMeter]
})
model_config = config.SenseModelParameters(
model_name=args.config_name,
hidden_size=args.hidden_size,
freeze_weights=args.freeze_weights,
context_layers=(-1, ))
configuration = config.Configuration(
model_parameters=model_config,
model=args.model,
save_path=args.save_path,
sequence_max_len=args.seq_len,
dropout_prob=args.dropout,
lr=args.lr,
batch_size=args.batch_size,
epochs=args.epochs,
device=torch.device(args.device),
tokenizer=transformers.AutoTokenizer.from_pretrained(args.model),
)