task-t3

This model is a fine-tuned version of KpRT/task-t2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3985
• F1: 0.7669
• Chronic Disease F1: 0.7755
• Chronic Disease Num: 2507
• Cancer F1: 0.7152
• Cancer Num: 753
• Allergy F1: 0.7833
• Allergy Num: 271
• Treatment F1: 0.7715
• Treatment Num: 2963
• Other F1: 0
• Other Num: 0

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
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Chronic Disease F1	Chronic Disease Num	Cancer F1	Cancer Num	Allergy F1	Allergy Num	Treatment F1	Treatment Num	Other F1	Other Num
0.236	0.3135	100	0.4006	0.7553	0.7577	2507	0.7027	753	0.7767	271	0.7650	2963	0	0
0.2141	0.6270	200	0.4293	0.7513	0.7662	2507	0.6703	753	0.7850	271	0.7577	2963	0	0
0.2483	0.9404	300	0.4024	0.7628	0.7710	2507	0.6994	753	0.7765	271	0.7712	2963	0	0
0.19	1.2539	400	0.4005	0.7666	0.7764	2507	0.7010	753	0.8069	271	0.7720	2963	0	0
0.1997	1.5674	500	0.3986	0.7698	0.7786	2507	0.7137	753	0.7946	271	0.7750	2963	0	0
0.2297	1.8809	600	0.3985	0.7669	0.7755	2507	0.7152	753	0.7833	271	0.7715	2963	0	0

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1