Modelpublic

LiukG/mus_promoter-finetuned-lora-bert-base-lastln-t2t

sourceHugging Faceupdated 3y agoView on Hugging Face

0likes7downloads

Model Card

mus_promoter-finetuned-lora-bert-base-lastln-t2t

This model is a fine-tuned version of AIRI-Institute/gena-lm-bert-base-lastln-t2t on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.0934
—F1: 0.9867
—Mcc Score: 0.9683
—Accuracy: 0.9844

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: 0.0005
—trainbatchsize: 8
—evalbatchsize: 64
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Mcc Score	Accuracy
0.3791	0.43	100	0.2575	0.9296	0.8460	0.9219
0.1329	0.85	200	0.0857	0.9730	0.9359	0.9688
0.1085	1.28	300	0.1337	0.9730	0.9359	0.9688
0.0609	1.71	400	0.1476	0.9730	0.9359	0.9688
0.041	2.14	500	0.0040	1.0	1.0	1.0
0.0165	2.56	600	0.0766	0.9863	0.9686	0.9844
0.0152	2.99	700	0.0908	0.9867	0.9683	0.9844
0.0087	3.42	800	0.0940	0.9730	0.9359	0.9688
0.0156	3.85	900	0.0911	0.9867	0.9683	0.9844
0.0081	4.27	1000	0.0934	0.9867	0.9683	0.9844

Framework versions

—Transformers 4.37.2
—Pytorch 2.1.0+cu121
—Datasets 2.17.1
—Tokenizers 0.15.2