Variome_0.0001_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: 0.1367
• Precision: 0.6567
• Recall: 0.3308
• F1: 0.44
• Accuracy: 0.9842

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: 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
0.4296	5.0	25	0.1570	0.0	0.0	0.0	0.9794
0.1679	10.0	50	0.1546	0.0	0.0	0.0	0.9794
0.1632	15.0	75	0.1375	0.0	0.0	0.0	0.9794
0.1332	20.0	100	0.1357	0.2381	0.0376	0.0649	0.9799
0.0886	25.0	125	0.1250	0.2222	0.0602	0.0947	0.9805
0.0714	30.0	150	0.1278	0.3333	0.1053	0.16	0.9809
0.0479	35.0	175	0.1220	0.5	0.2256	0.3109	0.9831
0.0301	40.0	200	0.1259	0.6154	0.3008	0.4040	0.9841
0.0198	45.0	225	0.1257	0.6364	0.3158	0.4221	0.9846
0.0138	50.0	250	0.1240	0.6184	0.3534	0.4498	0.9847
0.0099	55.0	275	0.1301	0.5823	0.3459	0.4340	0.9837
0.008	60.0	300	0.1343	0.5584	0.3233	0.4095	0.9832
0.0066	65.0	325	0.1290	0.5625	0.3383	0.4225	0.9830
0.0054	70.0	350	0.1366	0.6061	0.3008	0.4020	0.9838
0.0047	75.0	375	0.1334	0.6111	0.3308	0.4293	0.9841
0.0044	80.0	400	0.1367	0.6567	0.3308	0.44	0.9842

Framework versions

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