MultiCorp_all_label_5e-05_0404_ES2_strict_tok

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: 0.1346
• Precision: 0.3090
• Recall: 0.1750
• F1: 0.2235
• Accuracy: 0.9674

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: 2000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.9281	0.08	25	0.2509	0.0	0.0	0.0	0.9637
0.2583	0.15	50	0.2399	0.0	0.0	0.0	0.9637
0.2319	0.23	75	0.2011	0.0	0.0	0.0	0.9637
0.1901	0.31	100	0.1717	0.3333	0.0014	0.0028	0.9639
0.1894	0.39	125	0.1740	0.0	0.0	0.0	0.9637
0.1492	0.46	150	0.1454	0.1955	0.0320	0.0550	0.9635
0.1504	0.54	175	0.1437	0.1288	0.0139	0.0251	0.9643
0.1559	0.62	200	0.1326	0.1795	0.1123	0.1382	0.9665
0.1571	0.7	225	0.1406	0.3095	0.0604	0.1010	0.9613
0.1353	0.77	250	0.1346	0.3090	0.1750	0.2235	0.9674

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu118
• Datasets 2.11.0
• Tokenizers 0.13.3