metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: hubert-large-ls960-ft-finetuned-gtzan
results: []
hubert-large-ls960-ft-finetuned-gtzan
This model is a fine-tuned version of facebook/hubert-large-ls960-ft on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.8531
- Accuracy: 0.78
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: 2e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- 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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.279 | 1.0 | 112 | 2.2924 | 0.08 |
| 2.0905 | 2.0 | 225 | 2.0931 | 0.31 |
| 1.7503 | 3.0 | 337 | 1.6857 | 0.46 |
| 1.7021 | 4.0 | 450 | 1.5041 | 0.54 |
| 1.3693 | 5.0 | 562 | 1.4453 | 0.53 |
| 1.1515 | 6.0 | 675 | 1.2720 | 0.62 |
| 1.0195 | 7.0 | 787 | 1.2036 | 0.61 |
| 0.8957 | 8.0 | 900 | 1.1265 | 0.62 |
| 0.9654 | 9.0 | 1012 | 1.0117 | 0.68 |
| 1.0166 | 10.0 | 1125 | 0.9691 | 0.68 |
| 0.8868 | 11.0 | 1237 | 1.0249 | 0.69 |
| 0.8822 | 12.0 | 1350 | 0.9859 | 0.69 |
| 0.808 | 13.0 | 1462 | 0.8248 | 0.75 |
| 0.7107 | 14.0 | 1575 | 0.9660 | 0.71 |
| 0.7964 | 15.0 | 1687 | 0.8939 | 0.73 |
| 0.69 | 16.0 | 1800 | 0.8490 | 0.75 |
| 0.4321 | 17.0 | 1912 | 0.8282 | 0.77 |
| 0.4942 | 18.0 | 2025 | 0.8220 | 0.78 |
| 0.4275 | 19.0 | 2137 | 0.8435 | 0.79 |
| 0.4661 | 19.91 | 2240 | 0.8531 | 0.78 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3