Kikia26/FineTunePubMedBertWithTensorflowKeras2

This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.0693
• Validation Loss: 0.3774
• Train Precision: 0.6399
• Train Recall: 0.7384
• Train F1: 0.6856
• Train Accuracy: 0.9030
• Epoch: 19

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:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-05, 'decay_steps': 200, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: float32

Training results

Train Loss	Validation Loss	Train Precision	Train Recall	Train F1	Train Accuracy	Epoch
1.5823	0.9047	0.0	0.0	0.0	0.7808	0
0.9053	0.6998	0.5303	0.0738	0.1296	0.8106	1
0.6980	0.5341	0.7038	0.3861	0.4986	0.8591	2
0.5206	0.4613	0.6213	0.5295	0.5718	0.8753	3
0.4110	0.4201	0.6292	0.5549	0.5897	0.8836	4
0.3260	0.3918	0.6306	0.5907	0.6100	0.8937	5
0.2682	0.3682	0.5989	0.6709	0.6328	0.8985	6
0.2240	0.3445	0.6355	0.6730	0.6537	0.9041	7
0.1891	0.3593	0.5736	0.7152	0.6366	0.8913	8
0.1672	0.3609	0.5721	0.7278	0.6407	0.8908	9
0.1456	0.3594	0.5940	0.7131	0.6481	0.8969	10
0.1310	0.3519	0.6437	0.7089	0.6747	0.9073	11
0.1103	0.3531	0.6322	0.7215	0.6739	0.9030	12
0.1014	0.3814	0.6065	0.7511	0.6711	0.8964	13
0.0945	0.3668	0.6494	0.7384	0.6910	0.9049	14
0.0880	0.3704	0.6510	0.7321	0.6892	0.9038	15
0.0836	0.3762	0.6377	0.7426	0.6862	0.9001	16
0.0709	0.3765	0.6354	0.7426	0.6848	0.9020	17
0.0755	0.3791	0.6347	0.7405	0.6835	0.9022	18
0.0693	0.3774	0.6399	0.7384	0.6856	0.9030	19

Framework versions

• Transformers 4.35.2
• TensorFlow 2.14.0
• Datasets 2.15.0
• Tokenizers 0.15.0