Yepes_5e-05_250

This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1394
• Precision: 0.7129
• Recall: 0.5498
• F1: 0.6208
• Accuracy: 0.9796

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: 5e-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
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.5163	1.39	25	0.2117	0.0	0.0	0.0	0.9672
0.1988	2.78	50	0.2076	0.0	0.0	0.0	0.9672
0.1579	4.17	75	0.1379	0.4017	0.2338	0.2956	0.9712
0.1055	5.56	100	0.1182	0.5688	0.3085	0.4	0.9754
0.0791	6.94	125	0.1024	0.5032	0.3955	0.4429	0.9762
0.0545	8.33	150	0.1038	0.5683	0.4453	0.4993	0.9777
0.0402	9.72	175	0.1165	0.7063	0.4726	0.5663	0.9796
0.0337	11.11	200	0.1104	0.6635	0.5149	0.5798	0.9786
0.0238	12.5	225	0.1203	0.6789	0.5522	0.6091	0.9790
0.0202	13.89	250	0.1263	0.7416	0.5498	0.6314	0.9803
0.0147	15.28	275	0.1273	0.6965	0.5423	0.6098	0.9791
0.0129	16.67	300	0.1338	0.6796	0.5647	0.6168	0.9787
0.0109	18.06	325	0.1359	0.7690	0.5547	0.6445	0.9804
0.0091	19.44	350	0.1394	0.7129	0.5498	0.6208	0.9796

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• Tokenizers 0.13.2