README.md

metadata

license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - generated_from_trainer
datasets:
  - marsyas/gtzan
metrics:
  - accuracy
model-index:
  - name: HamzaSidhu786/distilhubert-finetuned-gtzan
    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: 0.83

HamzaSidhu786/distilhubert-finetuned-gtzan

This model is a fine-tuned version of facebook/wav2vec2-base on the GTZAN dataset. It achieves the following results on the evaluation set:

• Loss: 0.8269
• 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: 16
• eval_batch_size: 16
• 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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1788	1.0	57	2.0907	0.39
1.6561	2.0	114	1.5747	0.62
1.3464	3.0	171	1.4279	0.57
1.1727	4.0	228	1.1862	0.68
0.9399	5.0	285	1.0572	0.66
0.931	6.0	342	1.1268	0.66
0.7375	7.0	399	0.8744	0.77
0.5798	8.0	456	0.8596	0.78
0.5668	9.0	513	0.8253	0.76
0.4972	10.0	570	0.8273	0.76
0.2375	11.0	627	0.8192	0.76
0.1913	12.0	684	0.7618	0.83
0.2132	13.0	741	0.8249	0.82
0.0823	14.0	798	0.8962	0.81
0.0444	15.0	855	0.9376	0.78
0.0375	16.0	912	0.8609	0.81
0.0298	17.0	969	0.8741	0.83
0.0808	18.0	1026	0.8911	0.84
0.0453	19.0	1083	0.8756	0.84
0.0229	20.0	1140	0.8269	0.83

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1