Add training script

train.py

@@ -0,0 +1,60 @@
+from datasets import load_dataset
+from span_marker import SpanMarkerModel, Trainer
+from transformers import TrainingArguments
+
+
+def main() -> None:
+    # Load the dataset, ensure "tokens" and "ner_tags" columns, and get a list of labels
+    dataset = load_dataset("DFKI-SLT/few-nerd", "supervised")
+    dataset = dataset.remove_columns("ner_tags")
+    dataset = dataset.rename_column("fine_ner_tags", "ner_tags")
+    labels = dataset["train"].features["ner_tags"].feature.names
+
+    # Initialize a SpanMarker model using a pretrained BERT-style encoder
+    model_name = "roberta-large"
+    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="models/span_marker_roberta_large_fewnerd_fine_super",
+        # Training Hyperparameters:
+        learning_rate=1e-5,
+        per_device_train_batch_size=8,
+        per_device_eval_batch_size=8,
+        num_train_epochs=3,
+        weight_decay=0.01,
+        warmup_ratio=0.1,
+        bf16=True,
+        # Other Training parameters
+        logging_first_step=True,
+        logging_steps=50,
+        evaluation_strategy="steps",
+        save_strategy="steps",
+        eval_steps=3000,
+        dataloader_num_workers=2,
+    )
+
+    # Initialize the trainer using our model, training args & dataset, and train
+    trainer = Trainer(
+        model=model,
+        args=args,
+        train_dataset=dataset["train"],
+        eval_dataset=dataset["validation"],
+    )
+    trainer.train()
+    trainer.save_model("models/span_marker_roberta_large_fewnerd_fine_super/checkpoint-final")
+
+    # Compute & save the metrics on the test set
+    metrics = trainer.evaluate(dataset["test"], metric_key_prefix="test")
+    trainer.save_metrics("test", metrics)
+
+
+if __name__ == "__main__":
+    main()