ICU_Returns_BioClinicalBERT

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: 1.7775
• F1:: 0.7063
• Roc Auc: 0.7198
• Precision with 0:: 0.8846
• Precision with 1:: 0.6538
• Recall with 0:: 0.5055
• Recal with 1:: 0.9341
• Accuracy:: 0.7198

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 13

Training results

Training Loss	Epoch	Step	Validation Loss	F1:	Roc Auc	Precision with 0:	Precision with 1:	Recall with 0:	Recal with 1:	Accuracy:
No log	1.0	46	0.6964	0.3573	0.5110	1.0	0.5056	0.0220	1.0	0.5110
No log	2.0	92	0.6611	0.5248	0.5714	0.6912	0.5439	0.2582	0.8846	0.5714
No log	3.0	138	0.6322	0.6315	0.6374	0.6838	0.6096	0.5110	0.7637	0.6374
No log	4.0	184	0.6526	0.6396	0.6566	0.7767	0.6092	0.4396	0.8736	0.6566
No log	5.0	230	0.6826	0.6693	0.6923	0.9070	0.6259	0.4286	0.9560	0.6923
No log	6.0	276	0.7496	0.7230	0.7335	0.8829	0.6680	0.5385	0.9286	0.7335
No log	7.0	322	1.5500	0.6398	0.6703	0.9079	0.6076	0.3791	0.9615	0.6703
No log	8.0	368	0.9037	0.7438	0.7527	0.9035	0.684	0.5659	0.9396	0.7527
No log	9.0	414	1.6723	0.6965	0.7143	0.9149	0.6444	0.4725	0.9560	0.7143
No log	10.0	460	1.4913	0.7030	0.7170	0.8835	0.6513	0.5	0.9341	0.7170
0.3158	11.0	506	1.7129	0.6990	0.7143	0.89	0.6477	0.4890	0.9396	0.7143
0.3158	12.0	552	1.8420	0.6882	0.7060	0.8947	0.6394	0.4670	0.9451	0.7060
0.3158	13.0	598	1.7775	0.7063	0.7198	0.8846	0.6538	0.5055	0.9341	0.7198

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.14.1