kinkpunk's picture
Update README.md
85809ea
metadata
license: bsd-3-clause
base_model: MIT/ast-finetuned-audioset-10-10-0.4593
tags:
  - generated_from_trainer
datasets:
  - marsyas/gtzan
metrics:
  - accuracy
model-index:
  - name: ast-finetuned-audioset-10-10-0.4593-finetuned-gtzan
    results: []

ast-finetuned-audioset-10-10-0.4593-finetuned-gtzan

This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on the GTZAN dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5270
  • Accuracy: 0.88

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: 4
  • eval_batch_size: 4
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.3141 1.0 112 0.6444 0.8
0.3915 2.0 225 0.5292 0.85
0.2685 3.0 337 0.4638 0.85
0.0043 4.0 450 0.6904 0.88
0.124 5.0 562 0.5522 0.91
0.0002 6.0 675 0.4958 0.87
0.0003 7.0 787 0.5430 0.87
0.0001 8.0 900 0.5116 0.89
0.1202 9.0 1012 0.5194 0.88
0.0001 9.96 1120 0.5270 0.88

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.13.1
  • Tokenizers 0.13.3