Modelpublic

lejonck/xlsr53-ptbr-mupe-final3

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes3downloads

Model Card

xlsr53-ptbr-mupe-final3

This model is a fine-tuned version of lejonck/xlsr53-ptbr-mupe-final2 on the None dataset. It achieves the following results on the evaluation set:

—Loss: 0.4183
—Wer: 0.1598
—Cer: 0.0626

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: 1e-05
—trainbatchsize: 2
—evalbatchsize: 2
—seed: 42
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
—lrschedulertype: linear
—lrschedulerwarmup_steps: 100
—num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
0.7433	1.0	2000	0.4015	0.1761	0.0662
0.7252	2.0	4000	0.3959	0.1732	0.0658
0.553	3.0	6000	0.3915	0.1643	0.0635
0.7068	4.0	8000	0.3978	0.1632	0.0643
0.4559	5.0	10000	0.4137	0.1624	0.0626
0.7618	6.0	12000	0.4074	0.1611	0.0620
0.4847	7.0	14000	0.4151	0.1605	0.0618
0.2968	8.0	16000	0.4183	0.1596	0.0625
0.3977	9.0	18000	0.4310	0.1619	0.0623
0.3833	10.0	20000	0.4314	0.1617	0.0620
0.5201	11.0	22000	0.4228	0.1613	0.0623
0.4379	12.0	24000	0.4298	0.1605	0.0622

Framework versions

—Transformers 4.55.1
—Pytorch 2.7.0+cu126
—Datasets 4.0.0
—Tokenizers 0.21.4