nlp_te_ner_biobert

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

• Loss: 0.0375
• Precision: 0.9570
• Recall: 0.9672
• F1: 0.9621
• Accuracy: 0.9900

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	0.0415	0.9548	0.9657	0.9602	0.9892
No log	2.0	42	0.0390	0.9551	0.9672	0.9611	0.9898
No log	3.0	63	0.0397	0.9515	0.9662	0.9588	0.9892
No log	4.0	84	0.0375	0.9570	0.9672	0.9621	0.9900
No log	5.0	105	0.0370	0.9560	0.9654	0.9607	0.9898
No log	6.0	126	0.0421	0.9471	0.9634	0.9552	0.9878
No log	7.0	147	0.0444	0.9424	0.9583	0.9503	0.9864
No log	8.0	168	0.0464	0.9448	0.9641	0.9544	0.9870
No log	9.0	189	0.0444	0.9453	0.9609	0.9530	0.9868

Framework versions

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