jialicheng's picture
Upload folder using huggingface_hub
a186b1a verified
metadata
license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
  - audio-classification
  - generated_from_trainer
datasets:
  - superb
metrics:
  - accuracy
model-index:
  - name: superb_ks_42
    results:
      - task:
          name: Audio Classification
          type: audio-classification
        dataset:
          name: superb
          type: superb
          config: ks
          split: validation
          args: ks
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9833774639599883

superb_ks_42

This model is a fine-tuned version of facebook/hubert-base-ls960 on the superb dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0857
  • Accuracy: 0.9834

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
  • train_batch_size: 32
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.5858 1.0 1597 0.1746 0.9701
0.2268 2.0 3194 0.1038 0.9773
0.1937 3.0 4791 0.0887 0.9797
0.163 4.0 6388 0.0786 0.9826
0.1468 5.0 7985 0.0982 0.9815
0.1327 6.0 9582 0.0888 0.9821
0.1175 7.0 11179 0.0967 0.9812
0.1164 8.0 12776 0.0922 0.9815
0.1023 9.0 14373 0.0875 0.9828
0.0983 10.0 15970 0.0857 0.9834

Framework versions

  • Transformers 4.40.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1