CoolFace
Modelpublic

Phakphum/outs

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
0likes7downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

outs

This model is a fine-tuned version of biodatlab/whisper-th-small-combined on an unknown dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.1631
  • —Cer: 10.9836

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: 5e-05
  • —trainbatchsize: 16
  • —evalbatchsize: 16
  • —seed: 42
  • —gradientaccumulationsteps: 2.0
  • —totaltrainbatch_size: 32.0
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 5.0
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossCer
No log0.9926670.187610.2687
0.12492.01350.163213.0653
0.07152.99262020.187014.0862
0.07154.02700.163610.3119
0.03864.96303350.163110.9836

Framework versions

  • —Transformers 4.44.2
  • —Pytorch 2.4.1+cu121
  • —Datasets 3.0.1
  • —Tokenizers 0.19.1