Edit model card

distilhubert-finetuned-gtzan-2

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

  • Loss: 0.7203
  • Accuracy: 0.86

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: 10
  • eval_batch_size: 10
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.2
  • num_epochs: 20
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.2521 1.0 90 2.2219 0.3
1.8502 2.0 180 1.8299 0.54
1.4155 3.0 270 1.4247 0.64
0.9885 4.0 360 1.0313 0.7
0.8111 5.0 450 0.8535 0.78
0.7023 6.0 540 0.7743 0.79
0.5663 7.0 630 0.6618 0.81
0.3577 8.0 720 0.6937 0.77
0.3003 9.0 810 0.6107 0.82
0.1321 10.0 900 0.5648 0.81
0.0488 11.0 990 0.5655 0.84
0.0323 12.0 1080 0.5612 0.86
0.0154 13.0 1170 0.6338 0.85
0.0108 14.0 1260 0.7292 0.84
0.0082 15.0 1350 0.7542 0.84
0.0065 16.0 1440 0.7123 0.86
0.0062 17.0 1530 0.6949 0.86
0.0848 18.0 1620 0.7332 0.85
0.0053 19.0 1710 0.7291 0.85
0.005 20.0 1800 0.7203 0.86

Framework versions

  • Transformers 4.38.1
  • Pytorch 2.2.1
  • Datasets 2.17.1
  • Tokenizers 0.15.2
Downloads last month
9
Safetensors
Model size
23.7M params
Tensor type
F32
·

Finetuned from

Dataset used to train SavorSauce/distilhubert-finetuned-gtzan-2

Evaluation results