finetuned_ClinicalLongformer_newData

This model is a fine-tuned version of yikuan8/Clinical-Longformer on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5642
• F1: 0.8806

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.5848	1.0	25	0.5680	0.8333
0.4637	2.0	50	0.5057	0.8333
0.3851	3.0	75	0.5439	0.8633
0.3559	4.0	100	0.5642	0.8806
0.2903	5.0	125	0.5273	0.875
0.3463	6.0	150	0.7850	0.8741
0.3455	7.0	175	0.6507	0.8788
0.1776	8.0	200	0.6855	0.8769
0.1864	9.0	225	0.9215	0.8593
0.1548	10.0	250	0.9744	0.8485
0.0926	11.0	275	1.0779	0.8485
0.0857	12.0	300	1.4990	0.8369
0.0822	13.0	325	1.2385	0.8485
0.0783	14.0	350	1.4154	0.8444
0.0876	15.0	375	1.4720	0.8444
0.0152	16.0	400	1.4362	0.8397
0.0079	17.0	425	1.6063	0.8406
0.027	18.0	450	1.6123	0.8382
0.0403	19.0	475	1.5967	0.8444
0.0046	20.0	500	1.6041	0.8444

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0