nlp_te_ner_scibert

This model is a fine-tuned version of AmedeoBonatti/nlp_te_mlm_scibert on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0064
• Precision: 0.9950
• Recall: 0.9955
• F1: 0.9952
• Accuracy: 0.9990

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: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	21	0.0062	0.9942	0.9952	0.9947	0.9989
No log	2.0	42	0.0064	0.9950	0.9955	0.9952	0.9990
No log	3.0	63	0.0067	0.9942	0.9957	0.9950	0.9988
No log	4.0	84	0.0067	0.9935	0.9957	0.9946	0.9988
No log	5.0	105	0.0095	0.9880	0.9925	0.9902	0.9974
No log	6.0	126	0.0068	0.9940	0.9952	0.9946	0.9986
No log	7.0	147	0.0065	0.9932	0.9945	0.9939	0.9987

Framework versions

• Transformers 4.41.2
• Pytorch 2.2.1
• Datasets 2.19.2
• Tokenizers 0.19.1