Edit model card

distilhubert-finetuned-gtzan

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: 0.8933
  • Accuracy: 0.83

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: 0.1
  • num_epochs: 15

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.1059 1.0 113 1.9709 0.39
1.4561 2.0 226 1.2865 0.59
1.03 3.0 339 0.9918 0.75
0.8979 4.0 452 0.8700 0.77
0.6697 5.0 565 0.7090 0.79
0.3289 6.0 678 0.6646 0.77
0.3612 7.0 791 0.6384 0.83
0.068 8.0 904 0.5989 0.85
0.1159 9.0 1017 0.7136 0.83
0.0228 10.0 1130 0.8329 0.84
0.0484 11.0 1243 0.8401 0.84
0.0283 12.0 1356 0.8522 0.84
0.008 13.0 1469 0.8865 0.84
0.0066 14.0 1582 0.9048 0.85
0.0067 15.0 1695 0.8933 0.83

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.0
  • Datasets 2.14.4
  • Tokenizers 0.13.3
Downloads last month
2
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Finetuned from

Dataset used to train reichenbach/distilhubert-finetuned-gtzan

Evaluation results