metadata
license: bsd-3-clause
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: ast_5-finetuned-ICBHI
results: []
ast_5-finetuned-ICBHI
This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2026
- Accuracy: 0.6547
- Sensitivity: 0.4720
- Specificity: 0.8181
- Score: 0.6451
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: 3e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Sensitivity | Specificity | Score |
---|---|---|---|---|---|---|---|
0.9246 | 1.0 | 258 | 0.9810 | 0.6076 | 0.5004 | 0.7035 | 0.6019 |
0.8279 | 2.0 | 517 | 0.8591 | 0.6478 | 0.3691 | 0.8970 | 0.6331 |
0.5969 | 3.0 | 776 | 0.9275 | 0.6446 | 0.5111 | 0.7639 | 0.6375 |
0.185 | 4.0 | 1035 | 1.2026 | 0.6547 | 0.4720 | 0.8181 | 0.6451 |
0.0305 | 4.99 | 1290 | 1.5138 | 0.6467 | 0.4873 | 0.7893 | 0.6383 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3