README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - marsyas/gtzan
metrics:
  - accuracy
model-index:
  - name: distilhubert-music-classification
    results:
      - task:
          name: Audio Classification
          type: audio-classification
        dataset:
          name: GTZAN
          type: marsyas/gtzan
          config: all
          split: train
          args: all
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.86

distilhubert-music-classification

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: 0.7110
• 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: 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: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1284	1.0	113	1.9802	0.5
1.435	2.0	226	1.3403	0.65
1.0235	3.0	339	0.9941	0.74
0.8973	4.0	452	0.9184	0.69
0.7312	5.0	565	0.6918	0.79
0.4306	6.0	678	0.6343	0.78
0.4204	7.0	791	0.6174	0.83
0.1326	8.0	904	0.5888	0.83
0.0766	9.0	1017	0.5939	0.84
0.0308	10.0	1130	0.7191	0.86
0.0318	11.0	1243	0.7308	0.84
0.0657	12.0	1356	0.7222	0.81
0.0096	13.0	1469	0.7075	0.84
0.0077	14.0	1582	0.7268	0.84
0.0073	15.0	1695	0.6957	0.85
0.0066	16.0	1808	0.7110	0.86

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3