Modelpublic

DanishMahdi/shah-latif-melody-classifier

sourceHugging Facemitupdated 1y agoView on Hugging Face

0likes57downloads

Model Card

results

This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

—Loss: 1.7003
—Accuracy: 0.5660
—F1: 0.5571

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: 8
—evalbatchsize: 8
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
—lrschedulertype: linear
—num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	477	2.8092	0.2442	0.1739
2.961	2.0	954	2.4814	0.3187	0.2486
2.6582	3.0	1431	2.2660	0.3889	0.3370
2.3599	4.0	1908	2.0642	0.4277	0.3836
2.0474	5.0	2385	1.9820	0.4476	0.4212
1.7898	6.0	2862	1.9047	0.4738	0.4416
1.598	7.0	3339	1.7972	0.4937	0.4680
1.4538	8.0	3816	1.7815	0.5063	0.4805
1.2541	9.0	4293	1.7712	0.5199	0.4941
1.1149	10.0	4770	1.7398	0.5189	0.4950
0.9883	11.0	5247	1.7048	0.5304	0.5185
0.9024	12.0	5724	1.7161	0.5503	0.5390
0.7866	13.0	6201	1.6983	0.5535	0.5414
0.721	14.0	6678	1.7003	0.5660	0.5571
0.6566	15.0	7155	1.6980	0.5556	0.5492

Framework versions

—Transformers 4.56.1
—Pytorch 2.8.0+cu126
—Datasets 4.0.0
—Tokenizers 0.22.0