ClinicalBERT-ft-text_cl

This model is a fine-tuned version of medicalai/ClinicalBERT on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0173
• Accuracy: 0.8880
• Recall: 0.9412
• Precision: 0.8766
• F1: 0.9078

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1
No log	1.0	231	0.3258	0.8658	0.9552	0.8383	0.8929
No log	2.0	462	0.4387	0.8664	0.9711	0.8298	0.8949
0.2029	3.0	693	0.4922	0.8786	0.9771	0.8413	0.9041
0.2029	4.0	924	0.6135	0.8874	0.9582	0.8643	0.9088
0.0562	5.0	1155	0.6957	0.8821	0.9592	0.8568	0.9051
0.0562	6.0	1386	0.8482	0.8734	0.9612	0.8443	0.8989
0.0169	7.0	1617	0.6741	0.9008	0.9542	0.8854	0.9185
0.0169	8.0	1848	0.8648	0.8641	0.9631	0.8315	0.8925
0.0132	9.0	2079	0.7180	0.8967	0.9313	0.8965	0.9135
0.0132	10.0	2310	0.8567	0.8891	0.9552	0.8687	0.9099
0.0044	11.0	2541	0.7808	0.9002	0.9293	0.9032	0.9161
0.0044	12.0	2772	1.0333	0.8629	0.9671	0.8278	0.8921
0.0033	13.0	3003	0.9561	0.8839	0.9552	0.8616	0.9060
0.0033	14.0	3234	0.8847	0.8897	0.9522	0.8715	0.9100
0.0033	15.0	3465	1.2981	0.8471	0.9831	0.8011	0.8828
0.0028	16.0	3696	1.0164	0.8722	0.9542	0.8470	0.8974
0.0028	17.0	3927	0.8194	0.8915	0.9512	0.8745	0.9113
0.0114	18.0	4158	0.8303	0.9014	0.9502	0.8891	0.9186
0.0114	19.0	4389	0.9197	0.8874	0.9562	0.8656	0.9087
0.0029	20.0	4620	1.0221	0.8845	0.9592	0.8598	0.9068
0.0029	21.0	4851	0.9183	0.8926	0.9363	0.8868	0.9109
0.0026	22.0	5082	1.0870	0.8816	0.9572	0.8573	0.9045
0.0026	23.0	5313	1.0594	0.8856	0.9542	0.8646	0.9072
0.0025	24.0	5544	0.9493	0.8973	0.9353	0.8943	0.9143
0.0025	25.0	5775	0.9936	0.8880	0.9502	0.8704	0.9086
0.0026	26.0	6006	1.0409	0.8886	0.9552	0.8679	0.9094
0.0026	27.0	6237	1.0020	0.8874	0.9452	0.8730	0.9077
0.0026	28.0	6468	1.0178	0.8868	0.9412	0.875	0.9069
0.0016	29.0	6699	1.0101	0.8868	0.9392	0.8764	0.9067
0.0016	30.0	6930	1.0173	0.8880	0.9412	0.8766	0.9078

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Tokenizers 0.21.1