ner_cs

This model is a fine-tuned version of google/mt5-large on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5017
• Loc: {'precision': 0.8522895125553914, 'recall': 0.9058084772370487, 'f1': 0.878234398782344, 'number': 637}
• Org: {'precision': 0.8361702127659575, 'recall': 0.8488120950323974, 'f1': 0.8424437299035369, 'number': 463}
• Per: {'precision': 0.9230769230769231, 'recall': 0.9737470167064439, 'f1': 0.9477351916376306, 'number': 419}
• Overall Precision: 0.8672
• Overall Recall: 0.9072
• Overall F1: 0.8867
• Overall Accuracy: 0.9365

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Loc	Org	Per	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.2192	5.71	5000	0.2824	{'precision': 0.8384728340675477, 'recall': 0.8963893249607535, 'f1': 0.8664643399089529, 'number': 637}	{'precision': 0.808641975308642, 'recall': 0.8488120950323974, 'f1': 0.8282402528977871, 'number': 463}	{'precision': 0.9325581395348838, 'recall': 0.9570405727923628, 'f1': 0.944640753828033, 'number': 419}	0.8547	0.8986	0.8761	0.9363
0.0244	11.43	10000	0.4134	{'precision': 0.8622754491017964, 'recall': 0.9042386185243328, 'f1': 0.8827586206896552, 'number': 637}	{'precision': 0.841991341991342, 'recall': 0.8401727861771058, 'f1': 0.8410810810810811, 'number': 463}	{'precision': 0.920814479638009, 'recall': 0.9713603818615751, 'f1': 0.9454123112659697, 'number': 419}	0.8728	0.9032	0.8877	0.9370
0.0066	17.14	15000	0.5017	{'precision': 0.8522895125553914, 'recall': 0.9058084772370487, 'f1': 0.878234398782344, 'number': 637}	{'precision': 0.8361702127659575, 'recall': 0.8488120950323974, 'f1': 0.8424437299035369, 'number': 463}	{'precision': 0.9230769230769231, 'recall': 0.9737470167064439, 'f1': 0.9477351916376306, 'number': 419}	0.8672	0.9072	0.8867	0.9365

Framework versions

• Transformers 4.39.3
• Pytorch 1.11.0a0+17540c5
• Datasets 2.20.0
• Tokenizers 0.15.2