Bio_ClinicalBERT_fold_10_binary_v1

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: 1.5504
• F1: 0.8243

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	288	0.3803	0.8103
0.4005	2.0	576	0.4769	0.8070
0.4005	3.0	864	0.5258	0.7955
0.1889	4.0	1152	0.7423	0.8153
0.1889	5.0	1440	1.1246	0.8012
0.0703	6.0	1728	1.1325	0.8039
0.0246	7.0	2016	1.2192	0.8196
0.0246	8.0	2304	1.3645	0.8050
0.0192	9.0	2592	1.4029	0.8087
0.0192	10.0	2880	1.3714	0.8117
0.0107	11.0	3168	1.4673	0.8092
0.0107	12.0	3456	1.3941	0.8199
0.0084	13.0	3744	1.4350	0.8126
0.0083	14.0	4032	1.4428	0.8162
0.0083	15.0	4320	1.2892	0.8263
0.0119	16.0	4608	1.4238	0.8222
0.0119	17.0	4896	1.4961	0.8174
0.0046	18.0	5184	1.5010	0.8107
0.0046	19.0	5472	1.4876	0.8215
0.0036	20.0	5760	1.5080	0.8180
0.0031	21.0	6048	1.5317	0.8261
0.0031	22.0	6336	1.5103	0.8215
0.0005	23.0	6624	1.5255	0.8197
0.0005	24.0	6912	1.5578	0.8257
0.0001	25.0	7200	1.5504	0.8243

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1