tomaarsen
/

span-marker-bert-base-cross-ner

Token Classification

named-entity-recognition

Model card Files Files and versions Metrics Training metrics Community

span-marker-bert-base-cross-ner / train.py

tomaarsen's picture

tomaarsen HF staff

Upload train.py with huggingface_hub

2bc8e1c 9 months ago

raw history blame contribute delete

No virus

2.07 kB

	from datasets import load_dataset
	from transformers import TrainingArguments
	from span_marker import SpanMarkerModel, Trainer


	def main() -> None:
	# Load the dataset, ensure "tokens" and "ner_tags" columns, and get a list of labels
	train_dataset = load_dataset("P3ps/Cross_ner", split="train")
	test_dataset = load_dataset("P3ps/Cross_ner", split="test")
	labels = train_dataset.features["ner_tags"].feature.names

	# Initialize a SpanMarker model using a pretrained BERT-style encoder
	model_name = "bert-base-cased"
	model = SpanMarkerModel.from_pretrained(
	model_name,
	labels=labels,
	# SpanMarker hyperparameters:
	model_max_length=256,
	marker_max_length=128,
	entity_max_length=8,
	)

	# Prepare the 🤗 transformers training arguments
	args = TrainingArguments(
	output_dir=f"models/span_marker_bert_base_cross_ner",
	run_name=f"bb_cross_ner",
	# Training Hyperparameters:
	learning_rate=5e-5,
	per_device_train_batch_size=32,
	per_device_eval_batch_size=32,
	num_train_epochs=3,
	weight_decay=0.01,
	warmup_ratio=0.1,
	bf16=True, # Replace `bf16` with `fp16` if your hardware can't use bf16.
	# Other Training parameters
	logging_first_step=True,
	logging_steps=50,
	evaluation_strategy="steps",
	save_strategy="steps",
	eval_steps=200,
	save_total_limit=2,
	dataloader_num_workers=2,
	)

	# Initialize the trainer using our model, training args & dataset, and train
	trainer = Trainer(
	model=model,
	args=args,
	train_dataset=train_dataset,
	eval_dataset=test_dataset,
	)
	trainer.train()
	trainer.save_model(f"models/span_marker_bert_base_cross_ner/checkpoint-final")

	# Compute & save the metrics on the test set
	metrics = trainer.evaluate(test_dataset, metric_key_prefix="test")
	trainer.save_metrics("test", metrics)
	trainer.create_model_card()


	if __name__ == "__main__":
	main()