CoolFace
Modelpublic

Pageee/DistilFT-Frisian-10hd

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes3downloads
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. -->

DistilFT-Frisian-10h

This model is a fine-tuned version of distil-small.en on the mozilla-foundation/commonvoice6fyNL dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.5755
  • —Wer: 26.9114

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: 0.0001
  • —trainbatchsize: 8
  • —evalbatchsize: 8
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 300
  • —training_steps: 5000
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
0.95040.53485001.093957.4122
0.46561.069510000.824145.7316
0.45331.604315000.728541.3474
0.17452.139020000.687537.7009
0.17012.673825000.626134.7603
0.07093.208630000.656633.4415
0.07313.743335000.588030.5650
0.02344.278140000.594928.8754
0.01924.812845000.579927.7063
0.00385.347650000.575526.9114

Framework versions

  • —Transformers 4.41.0.dev0
  • —Pytorch 2.3.0+cu121
  • —Datasets 2.19.1
  • —Tokenizers 0.19.1