Modelpublic

Tommert25/multibert_dataaugmentation

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes11downloads

Model Card

multibert_dataaugmentation

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

—Loss: 0.7138
—Precisions: 0.8609
—Recall: 0.8356
—F-measure: 0.8464
—Accuracy: 0.8989

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: 7.5e-05
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 14

Training results

Training Loss	Epoch	Step	Validation Loss	Precisions	Recall	F-measure	Accuracy
0.5775	1.0	285	0.4827	0.7847	0.7040	0.7340	0.8509
0.2623	2.0	570	0.5829	0.8035	0.7359	0.7591	0.8613
0.1503	3.0	855	0.5609	0.7946	0.8083	0.7917	0.8804
0.088	4.0	1140	0.5481	0.8406	0.7997	0.8170	0.8860
0.0592	5.0	1425	0.6359	0.8207	0.8210	0.8120	0.8828
0.0414	6.0	1710	0.6589	0.8313	0.8171	0.8198	0.8843
0.0271	7.0	1995	0.7117	0.8689	0.7882	0.8216	0.8936
0.0179	8.0	2280	0.7138	0.8609	0.8356	0.8464	0.8989
0.0121	9.0	2565	0.7289	0.8456	0.8128	0.8278	0.8946
0.0081	10.0	2850	0.7603	0.8344	0.8223	0.8278	0.8956
0.0058	11.0	3135	0.8126	0.8576	0.8107	0.8322	0.8942
0.0041	12.0	3420	0.8004	0.8582	0.8267	0.8415	0.8955
0.0031	13.0	3705	0.7936	0.8599	0.8275	0.8426	0.8961
0.0028	14.0	3990	0.8076	0.8602	0.8226	0.8401	0.8966

Framework versions

—Transformers 4.34.0
—Pytorch 2.0.1+cu118
—Datasets 2.14.5
—Tokenizers 0.14.1