metadata
license: bsd-3-clause
base_model: MIT/ast-finetuned-audioset-10-10-0.450
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: ast-finetuned-audioset-10-10-0.450_ESC50
results: []
ast-finetuned-audioset-10-10-0.450_ESC50
This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.450 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2887
- Accuracy: 0.9275
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: 1e-05
- train_batch_size: 6
- eval_batch_size: 6
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 24
- 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 |
|---|---|---|---|---|
| 2.7745 | 0.99 | 66 | 2.3340 | 0.605 |
| 0.7521 | 1.99 | 133 | 0.8978 | 0.8875 |
| 0.2307 | 3.0 | 200 | 0.5545 | 0.8975 |
| 0.0903 | 4.0 | 267 | 0.4063 | 0.925 |
| 0.03 | 4.99 | 333 | 0.3488 | 0.92 |
| 0.0123 | 5.99 | 400 | 0.2987 | 0.925 |
| 0.0101 | 7.0 | 467 | 0.2887 | 0.9275 |
| 0.0067 | 8.0 | 534 | 0.2808 | 0.9275 |
| 0.0055 | 8.99 | 600 | 0.2784 | 0.9275 |
| 0.0051 | 9.89 | 660 | 0.2778 | 0.9275 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2