xlm-roberta-large-finetuned-ner-vlsp2021-3090-29June-1

This model is a fine-tuned version of FacebookAI/xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0723
• Atetime: {'precision': 0.8662733529990168, 'recall': 0.8792415169660679, 'f1': 0.8727092620108965, 'number': 1002}
• Ddress: {'precision': 0.78125, 'recall': 0.8620689655172413, 'f1': 0.8196721311475409, 'number': 29}
• Erson: {'precision': 0.9603217158176943, 'recall': 0.943127962085308, 'f1': 0.9516471838469712, 'number': 1899}
• Ersontype: {'precision': 0.7422222222222222, 'recall': 0.7324561403508771, 'f1': 0.737306843267108, 'number': 684}
• Honenumber: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 9}
• Iscellaneous: {'precision': 0.5526315789473685, 'recall': 0.5283018867924528, 'f1': 0.5401929260450161, 'number': 159}
• Mail: {'precision': 1.0, 'recall': 0.9411764705882353, 'f1': 0.9696969696969697, 'number': 51}
• Ocation: {'precision': 0.8572496263079222, 'recall': 0.8816295157571099, 'f1': 0.8692686623721108, 'number': 1301}
• P: {'precision': 1.0, 'recall': 0.9090909090909091, 'f1': 0.9523809523809523, 'number': 11}
• Rl: {'precision': 0.7647058823529411, 'recall': 0.8666666666666667, 'f1': 0.8125, 'number': 15}
• Roduct: {'precision': 0.7094155844155844, 'recall': 0.6992, 'f1': 0.7042707493956486, 'number': 625}
• Overall Precision: 0.8559
• Overall Recall: 0.8550
• Overall F1: 0.8554
• Overall Accuracy: 0.9802

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Atetime	Ddress	Erson	Ersontype	Honenumber	Iscellaneous	Mail	Ocation	P	Rl	Roduct	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.0783	1.0	3263	0.0723	{'precision': 0.8662733529990168, 'recall': 0.8792415169660679, 'f1': 0.8727092620108965, 'number': 1002}	{'precision': 0.78125, 'recall': 0.8620689655172413, 'f1': 0.8196721311475409, 'number': 29}	{'precision': 0.9603217158176943, 'recall': 0.943127962085308, 'f1': 0.9516471838469712, 'number': 1899}	{'precision': 0.7422222222222222, 'recall': 0.7324561403508771, 'f1': 0.737306843267108, 'number': 684}	{'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 9}	{'precision': 0.5526315789473685, 'recall': 0.5283018867924528, 'f1': 0.5401929260450161, 'number': 159}	{'precision': 1.0, 'recall': 0.9411764705882353, 'f1': 0.9696969696969697, 'number': 51}	{'precision': 0.8572496263079222, 'recall': 0.8816295157571099, 'f1': 0.8692686623721108, 'number': 1301}	{'precision': 1.0, 'recall': 0.9090909090909091, 'f1': 0.9523809523809523, 'number': 11}	{'precision': 0.7647058823529411, 'recall': 0.8666666666666667, 'f1': 0.8125, 'number': 15}	{'precision': 0.7094155844155844, 'recall': 0.6992, 'f1': 0.7042707493956486, 'number': 625}	0.8559	0.8550	0.8554	0.9802

Framework versions

• Transformers 4.40.2
• Pytorch 2.3.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1