biobert_model

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

• Loss: 0.9645
• Accuracy: 0.8711
• F1: 0.8475

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: 1e-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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	334	0.6463	0.6897	0.7129
0.4503	2.0	668	0.3590	0.8651	0.8269
0.2715	3.0	1002	0.4549	0.8711	0.8252
0.2715	4.0	1336	0.6012	0.8681	0.8434
0.1335	5.0	1670	0.6307	0.8576	0.8313
0.0746	6.0	2004	0.7658	0.8636	0.8366
0.0746	7.0	2338	0.8658	0.8666	0.8436
0.0307	8.0	2672	0.8312	0.8711	0.8453
0.0148	9.0	3006	0.8922	0.8651	0.8421
0.0148	10.0	3340	0.8761	0.8726	0.8490
0.0128	11.0	3674	0.9329	0.8681	0.8462
0.0105	12.0	4008	0.9512	0.8666	0.8441
0.0105	13.0	4342	0.9553	0.8711	0.8475
0.0069	14.0	4676	0.9731	0.8681	0.8445
0.0046	15.0	5010	0.9645	0.8711	0.8475

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3