ummonk's picture
End of training
e2a9a95 verified
|
raw
history blame
2.46 kB
metadata
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
  - generated_from_trainer
datasets:
  - audiofolder
metrics:
  - accuracy
model-index:
  - name: distilhubert-finetuned-accents
    results:
      - task:
          name: Audio Classification
          type: audio-classification
        dataset:
          name: audiofolder
          type: audiofolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.2708333333333333

distilhubert-finetuned-accents

This model is a fine-tuned version of ntu-spml/distilhubert on the audiofolder dataset. It achieves the following results on the evaluation set:

  • Loss: 1.9466
  • Accuracy: 0.2708

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.8
  • num_epochs: 12
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.48 1.0 48 2.4777 0.1042
2.473 2.0 96 2.4604 0.1562
2.4772 3.0 144 2.4282 0.1042
2.3678 4.0 192 2.4007 0.1042
2.324 5.0 240 2.3261 0.2083
2.2489 6.0 288 2.2360 0.1771
1.9909 7.0 336 2.1544 0.1875
1.9903 8.0 384 2.0937 0.1875
2.0668 9.0 432 2.0222 0.2083
1.8473 10.0 480 2.0298 0.1875
1.8068 11.0 528 1.9965 0.25
1.699 12.0 576 1.9466 0.2708

Framework versions

  • Transformers 4.36.2
  • Pytorch 2.1.0+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0