Modelpublic

npark95/finetuned_ClinicalLongformer

sourceHugging Faceupdated 2y agoView on Hugging Face

0likes14downloads

Model Card

finetuned_ClinicalLongformer

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: 0.4698
—F1: 0.8598

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: 1.5e-05
—trainbatchsize: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.6611	1.0	21	0.5689	0.8067
0.5268	2.0	42	0.4698	0.8598
0.4322	3.0	63	0.5205	0.8485
0.3564	4.0	84	0.6989	0.7356
0.322	5.0	105	0.6610	0.7872
0.1975	6.0	126	1.0724	0.6824
0.2025	7.0	147	0.8713	0.8283
0.1605	8.0	168	0.9925	0.84
0.0834	9.0	189	1.2466	0.8235
0.0851	10.0	210	1.4596	0.8269
0.0742	11.0	231	1.4772	0.8269
0.0261	12.0	252	1.4519	0.8431
0.1153	13.0	273	1.4875	0.84
0.0016	14.0	294	1.4731	0.84
0.0036	15.0	315	1.5420	0.8431
0.0012	16.0	336	1.5392	0.8367
0.0092	17.0	357	1.4960	0.8515
0.0012	18.0	378	1.5128	0.8515
0.0007	19.0	399	1.5327	0.8515
0.0012	20.0	420	1.5213	0.8515

Framework versions

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