Bio_clinicalBERT-frozen_embeddings-30

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

• Loss: 1.2236
• Accuracy: 0.5153

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: 0.03
• 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
No log	1.0	115	2.2869	0.3712
No log	2.0	230	2.0574	0.3930
No log	3.0	345	3.3676	0.3886
No log	4.0	460	2.4398	0.3581
2.2747	5.0	575	1.8722	0.4410
2.2747	6.0	690	2.1077	0.3231
2.2747	7.0	805	1.9363	0.5109
2.2747	8.0	920	3.5410	0.2882
2.0667	9.0	1035	2.8723	0.2795
2.0667	10.0	1150	2.0155	0.4148
2.0667	11.0	1265	2.4170	0.4105
2.0667	12.0	1380	2.8414	0.3188
2.0667	13.0	1495	1.9712	0.4367
1.9756	14.0	1610	1.8535	0.4323
1.9756	15.0	1725	1.8735	0.5109
1.9756	16.0	1840	1.7002	0.4629
1.9756	17.0	1955	1.4065	0.4716
1.7396	18.0	2070	2.0771	0.3799
1.7396	19.0	2185	2.8953	0.3100
1.7396	20.0	2300	1.5128	0.4847
1.7396	21.0	2415	1.3819	0.4629
1.5953	22.0	2530	1.7965	0.4236
1.5953	23.0	2645	1.4167	0.4847
1.5953	24.0	2760	1.3443	0.4891
1.5953	25.0	2875	1.5971	0.4410
1.5953	26.0	2990	1.4340	0.4847
1.3758	27.0	3105	1.3012	0.5022
1.3758	28.0	3220	1.3027	0.4978
1.3758	29.0	3335	1.3025	0.4803
1.3758	30.0	3450	1.2236	0.5153

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Tokenizers 0.21.0