CoolFace
Modelpublic

Chenchuhui/Icelandic-noAug-to-withAug

sourceHugging Facecc-by-4.0updated 3y agoView on Hugging Face
0likes5downloads
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. -->

Icelandic-noAug-to-withAug

This model is a fine-tuned version of Chenchuhui/Icelandic-finetuned-no-data-augmentation on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.2435
  • —Wer: 0.2215

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: 0.0003
  • —trainbatchsize: 8
  • —evalbatchsize: 8
  • —seed: 42
  • —gradientaccumulationsteps: 2
  • —totaltrainbatch_size: 16
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 500
  • —num_epochs: 10

Training results

Training LossEpochStepValidation LossWer
No log0.2100.66730.4206
No log0.4200.53170.3602
No log0.6300.42290.3121
No log0.8400.37650.2819
0.78881.0500.34860.2606
0.78881.2600.30580.2394
0.78881.4700.29230.2371
0.78881.6800.27900.2271
0.78881.8900.25070.2215
0.37942.01000.23190.2114
0.37942.21100.21870.2002
0.37942.41200.23830.1991
0.37942.61300.23200.1980
0.37942.81400.23460.1980
0.26253.01500.22860.1935
0.26253.21600.23790.1924
0.26253.41700.21340.1857
0.26253.61800.21440.1890
0.26253.81900.22180.1890
0.22214.02000.21120.1935
0.22214.22100.20410.1890
0.22214.42200.19230.1745
0.22214.62300.20310.1913
0.22214.82400.22540.2036
0.17355.02500.22320.2025
0.17355.22600.19150.1812
0.17355.42700.21350.1946
0.17355.62800.22150.1969
0.17355.82900.22040.1991
0.18136.03000.22150.2047
0.18136.23100.22010.2002
0.18136.43200.23080.1991
0.18136.63300.24740.2069
0.18136.83400.23550.2081
0.20367.03500.22740.2047
0.20367.23600.23980.2081
0.20367.43700.21620.1957
0.20367.63800.22840.2047
0.20367.83900.24020.1980
0.19538.04000.19470.1946
0.19538.24100.21140.2069
0.19538.44200.23060.2237
0.19538.64300.21820.2170
0.19538.84400.20940.2170
0.19059.04500.26710.2315
0.19059.24600.31680.2136
0.19059.44700.30050.2248
0.19059.64800.24860.2260
0.19059.84900.23490.2047
0.20110.05000.24350.2215

Framework versions

  • —Transformers 4.35.2
  • —Pytorch 2.1.0+cu121
  • —Datasets 2.15.0
  • —Tokenizers 0.15.0