Yepes_2e-05_29_03

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: 0.1453
• Precision: 0.5461
• Recall: 0.4375
• F1: 0.4858
• Accuracy: 0.9769

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: 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
0.9751	5.0	25	0.2060	0.0	0.0	0.0	0.9697
0.1835	10.0	50	0.1590	0.0	0.0	0.0	0.9697
0.1271	15.0	75	0.1311	0.1447	0.125	0.1341	0.9691
0.093	20.0	100	0.1335	0.3043	0.1989	0.2405	0.9739
0.0739	25.0	125	0.1274	0.3615	0.2670	0.3072	0.9744
0.0554	30.0	150	0.1267	0.5	0.3239	0.3931	0.9761
0.0419	35.0	175	0.1283	0.4458	0.4205	0.4327	0.9753
0.0336	40.0	200	0.1343	0.4958	0.3352	0.4000	0.9756
0.0264	45.0	225	0.1314	0.5303	0.3977	0.4545	0.9770
0.022	50.0	250	0.1309	0.5468	0.4318	0.4825	0.9776
0.0185	55.0	275	0.1372	0.5468	0.4318	0.4825	0.9776
0.0163	60.0	300	0.1383	0.5315	0.4318	0.4765	0.9767
0.0146	65.0	325	0.1413	0.5486	0.4489	0.4937	0.9770
0.0131	70.0	350	0.1407	0.5781	0.4205	0.4868	0.9776
0.012	75.0	375	0.1428	0.5821	0.4432	0.5032	0.9776
0.011	80.0	400	0.1453	0.5461	0.4375	0.4858	0.9769

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2