xlm-roberta-base-finetuned-panx-it

This model is a fine-tuned version of xlm-roberta-base. It achieves the following results on the evaluation set:

• Loss: 0.2619
• F1 Score: 0.8321

Model description

This model is a fine-tuned version of xlm-roberta-base on the Italian subset of the PAN-X dataset for Named Entity Recognition (NER). The model has been fine-tuned to perform token classification tasks and is evaluated on its performance in identifying named entities in Italian text.

Intended uses & limitations

Intended uses:

Named Entity Recognition (NER) tasks specifically for Italian. Token classification tasks involving Italian text.

Limitations:

The model's performance is optimized for Italian and may not generalize well to other languages without further fine-tuning. The model's predictions are based on the data it was trained on and may not handle out-of-domain data as effectively.

Training and evaluation data

The model was fine-tuned on the Italian subset of the PAN-X dataset, which includes labeled examples of named entities in Italian text. The evaluation data is a separate portion of the same dataset, used to assess the model's performance.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 24
• eval_batch_size: 24
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Score
0.7217	1.0	70	0.3193	0.7343
0.2736	2.0	140	0.2760	0.8055
0.1838	3.0	210	0.2619	0.8321

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1