bert-base-NER-finetuned-ner-cerec

This model is a fine-tuned version of dslim/bert-base-NER on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1511
• Precision: 0.8652
• Recall: 0.8531
• F1: 0.8592
• Accuracy: 0.9790

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	45	0.4929	0.6532	0.5664	0.6067	0.9237
No log	2.0	90	0.3090	0.8145	0.7063	0.7566	0.9497
No log	3.0	135	0.2477	0.8433	0.7902	0.8159	0.9625
No log	4.0	180	0.2209	0.8169	0.8112	0.8140	0.9669
No log	5.0	225	0.1906	0.8369	0.8252	0.8310	0.9739
No log	6.0	270	0.1568	0.8662	0.8601	0.8632	0.9784
No log	7.0	315	0.1532	0.8732	0.8671	0.8702	0.9790
No log	8.0	360	0.1530	0.8671	0.8671	0.8671	0.9790
No log	9.0	405	0.1526	0.8723	0.8601	0.8662	0.9790
No log	10.0	450	0.1511	0.8652	0.8531	0.8592	0.9790

Framework versions

• Transformers 4.24.0
• Pytorch 1.12.1
• Datasets 2.10.1
• Tokenizers 0.11.0