Modelpublic

yicozy/study-dictionary-roberta-base

sourceHugging Facemitupdated 3y agoView on Hugging Face

0likes15downloads

Model Card

study-dictionary-roberta-base

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.0011
—F1: 1.0
—Roc Auc: 1.0
—Accuracy: 1.0
—Recall: 1.0

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
—trainbatchsize: 10
—evalbatchsize: 10
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—lrschedulerwarmup_ratio: 0.1
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy	Recall
0.3342	1.0	778	0.1192	0.0	0.5	0.0	0.0
0.1099	2.0	1556	0.1040	0.0	0.5	0.0	0.0
0.0892	3.0	2334	0.0465	0.6835	0.7644	0.5479	0.5293
0.0345	4.0	3112	0.0240	0.9147	0.9241	0.8817	0.8485
0.025	5.0	3890	0.0152	0.9594	0.9650	0.9493	0.9303
0.0144	6.0	4668	0.0114	0.9735	0.9811	0.9671	0.9625
0.0118	7.0	5446	0.0082	0.9779	0.9848	0.9717	0.9700
0.0081	8.0	6224	0.0057	0.9873	0.9887	0.9839	0.9774
0.0065	9.0	7002	0.0052	0.9839	0.9860	0.9848	0.9720
0.0054	10.0	7780	0.0039	0.9895	0.9904	0.9888	0.9809
0.0041	11.0	8558	0.0030	0.9942	0.9949	0.9925	0.9899
0.0036	12.0	9336	0.0026	0.9936	0.9940	0.9942	0.9881
0.0027	13.0	10114	0.0023	0.9956	0.9964	0.9958	0.9927
0.0023	14.0	10892	0.0018	0.9985	0.9986	0.9972	0.9972
0.0021	15.0	11670	0.0017	0.9985	0.9994	0.9974	0.9988
0.0018	16.0	12448	0.0015	0.9985	0.9992	0.9979	0.9985
0.0014	17.0	13226	0.0012	0.9997	0.9998	0.9994	0.9995
0.0013	18.0	14004	0.0011	1.0	1.0	1.0	1.0
0.0012	19.0	14782	0.0010	1.0	1.0	1.0	1.0
0.0012	20.0	15560	0.0010	1.0	1.0	1.0	1.0

Framework versions

—Transformers 4.33.2
—Pytorch 2.0.1+cu118
—Datasets 2.14.5
—Tokenizers 0.13.3