metadata
license: bsd-3-clause
tags:
- generated_from_trainer
datasets:
- audiofolder
metrics:
- f1
model-index:
- name: ast-finetuned-audioset-10-10-0.4593-finetuned-AST
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: audiofolder
type: audiofolder
config: Data_Train
split: train
args: Data_Train
metrics:
- name: F1
type: f1
value: 0.9263904420024879
ast-finetuned-audioset-10-10-0.4593-finetuned-AST
This model is a fine-tuned version of MIT/ast-finetuned-audioset-10-10-0.4593 on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.5280
- F1: 0.9264
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- 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 | F1 |
|---|---|---|---|---|
| 0.875 | 1.0 | 688 | 0.5758 | 0.8268 |
| 1.1367 | 2.0 | 1376 | 0.5628 | 0.8709 |
| 0.0831 | 3.0 | 2064 | 0.5236 | 0.8933 |
| 0.0964 | 4.0 | 2752 | 0.4338 | 0.9204 |
| 0.001 | 5.0 | 3440 | 0.5198 | 0.9174 |
| 0.0001 | 6.0 | 4128 | 0.5625 | 0.9150 |
| 0.0 | 7.0 | 4816 | 0.5280 | 0.9264 |
| 0.0 | 8.0 | 5504 | 0.5261 | 0.9221 |
| 0.0 | 9.0 | 6192 | 0.5256 | 0.9206 |
| 0.0 | 10.0 | 6880 | 0.5271 | 0.9193 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3