distil-whisper-large-v2-8-ls-finetuned-gtzan
This model is a fine-tuned version of rsonavane/distil-whisper-large-v2-8-ls on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.4346
- Accuracy: 0.91
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- 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 |
---|---|---|---|---|
0.7999 | 1.0 | 112 | 1.0678 | 0.6 |
0.7487 | 1.99 | 224 | 0.8678 | 0.7 |
0.4251 | 3.0 | 337 | 1.0246 | 0.65 |
0.081 | 4.0 | 449 | 0.4755 | 0.87 |
0.0487 | 4.99 | 561 | 0.5131 | 0.89 |
0.0027 | 6.0 | 674 | 0.4820 | 0.88 |
0.002 | 6.99 | 786 | 0.4103 | 0.91 |
0.0017 | 8.0 | 899 | 0.4562 | 0.92 |
0.0019 | 9.0 | 1011 | 0.4374 | 0.91 |
0.0011 | 9.97 | 1120 | 0.4346 | 0.91 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
- Downloads last month
- 6
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.