Yepes_2e-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: 0.1279
• Precision: 0.6833
• Recall: 0.5100
• F1: 0.5840
• Accuracy: 0.9788

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
0.8467	1.39	25	0.2149	0.0	0.0	0.0	0.9672
0.1988	2.78	50	0.1959	0.0	0.0	0.0	0.9672
0.156	4.17	75	0.1439	0.3268	0.2065	0.2530	0.9691
0.1128	5.56	100	0.1324	0.49	0.2438	0.3256	0.9730
0.0978	6.94	125	0.1222	0.4964	0.3433	0.4059	0.9747
0.0788	8.33	150	0.1154	0.5193	0.3682	0.4309	0.9760
0.067	9.72	175	0.1162	0.4711	0.3856	0.4241	0.9749
0.058	11.11	200	0.1236	0.5275	0.3582	0.4267	0.9761
0.0491	12.5	225	0.1177	0.4940	0.4104	0.4484	0.9754
0.0443	13.89	250	0.1235	0.5472	0.4179	0.4739	0.9767
0.0383	15.28	275	0.1198	0.5764	0.4502	0.5056	0.9770
0.0369	16.67	300	0.1219	0.5892	0.4602	0.5168	0.9776
0.0326	18.06	325	0.1261	0.7	0.4701	0.5625	0.9790
0.0305	19.44	350	0.1269	0.6904	0.4826	0.5681	0.9787
0.0269	20.83	375	0.1252	0.6656	0.5	0.5710	0.9783
0.025	22.22	400	0.1253	0.6529	0.5100	0.5726	0.9782
0.0244	23.61	425	0.1284	0.6875	0.4925	0.5739	0.9790
0.0224	25.0	450	0.1279	0.6833	0.5100	0.5840	0.9788

Framework versions

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