CoolFace
Modelpublic

gary109/ai-light-dance_stepmania_ft_wav2vec2-large-xlsr-53-v5

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
0likes21downloads
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. -->

ai-light-dancestepmaniaft_wav2vec2-large-xlsr-53-v5

This model is a fine-tuned version of gary109/ai-light-dance_stepmania_ft_wav2vec2-large-xlsr-53-v4 on the GARY109/AILIGHTDANCE - ONSET-STEPMANIA2 dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.0163
  • —Wer: 0.6622

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: 4e-05
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —gradientaccumulationsteps: 16
  • —totaltrainbatch_size: 64
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 100
  • —num_epochs: 10.0
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
0.88671.03761.03820.6821
0.88612.07521.02600.6686
0.86823.011281.03580.6604
0.86624.015041.02340.6665
0.84635.018801.03330.6666
0.85736.022561.01630.6622
0.86287.026321.02090.6551
0.84938.030081.05250.6582
0.83719.033841.04090.6515
0.822910.037601.05970.6523

Framework versions

  • —Transformers 4.21.0.dev0
  • —Pytorch 1.11.0+cu113
  • —Datasets 2.3.3.dev0
  • —Tokenizers 0.12.1