File size: 3,890 Bytes
3f90362 19896b0 3f90362 93cf0ee 3f90362 460f417 3f90362 ac71199 460f417 1a45862 ac71199 460f417 1a45862 ac71199 460f417 1a45862 ac71199 460f417 1a45862 ac71199 460f417 3f90362 1a45862 3f90362 460f417 3f90362 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 |
---
license: bsd-3-clause
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
base_model: MIT/ast-finetuned-audioset-10-10-0.4593
model-index:
- name: ast-finetuned-audioset-10-10-0.4593_ft_ESC-50_aug_0-1
results: []
---
# ast-finetuned-audioset-10-10-0.4593_ft_ESC-50_aug_0-1
This model is a fine-tuned version of [MIT/ast-finetuned-audioset-10-10-0.4593](https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593) on a subset of [ashraq/esc50](https://huggingface.co/datasets/ashraq/esc50) dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7391
- Accuracy: 0.9286
- Precision: 0.9449
- Recall: 0.9286
- F1: 0.9244
## Training and evaluation data
Training and evaluation data were augmented with audiomentations [GitHub: iver56/audiomentations](https://github.com/iver56/audiomentations) library and the following augmentation methods have been performed based on previous experiments [Elliott et al.: Tiny transformers for audio classification at the edge](https://arxiv.org/pdf/2103.12157.pdf):
**Gain**
- each audio sample is amplified/attenuated by a random factor between 0.5 and 1.5 with a 0.3 probability
**Noise**
- a random amount of Gaussian noise with a relative amplitude between 0.001 and 0.015 is added to each audio sample with a 0.5 probability
**Speed adjust**
- duration of each audio sample is extended by a random amount between 0.5 and 1.5 with a 0.3 probability
**Pitch shift**
- pitch of each audio sample is shifted by a random amount of semitones selected from the closed interval [-4,4] with a 0.3 probability
**Time masking**
- a random fraction of lenght of each audio sample in the range of (0,0.02] is erased with a 0.3 probability
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-06
- 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: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
| 9.9002 | 1.0 | 28 | 8.5662 | 0.0 | 0.0 | 0.0 | 0.0 |
| 5.7235 | 2.0 | 56 | 4.3990 | 0.0357 | 0.0238 | 0.0357 | 0.0286 |
| 2.4076 | 3.0 | 84 | 2.2972 | 0.4643 | 0.7405 | 0.4643 | 0.4684 |
| 1.4448 | 4.0 | 112 | 1.3975 | 0.7143 | 0.7340 | 0.7143 | 0.6863 |
| 0.8373 | 5.0 | 140 | 1.0468 | 0.8571 | 0.8524 | 0.8571 | 0.8448 |
| 0.7239 | 6.0 | 168 | 0.8518 | 0.8929 | 0.9164 | 0.8929 | 0.8766 |
| 0.6504 | 7.0 | 196 | 0.7391 | 0.9286 | 0.9449 | 0.9286 | 0.9244 |
| 0.535 | 8.0 | 224 | 0.6682 | 0.9286 | 0.9449 | 0.9286 | 0.9244 |
| 0.4237 | 9.0 | 252 | 0.6443 | 0.9286 | 0.9449 | 0.9286 | 0.9244 |
| 0.3709 | 10.0 | 280 | 0.6304 | 0.9286 | 0.9449 | 0.9286 | 0.9244 |
### Test results
| Parameter | Value |
|:------------------------:|:------------------:|
| test_loss | 0.5829914808273315 |
| test_accuracy | 0.9285714285714286 |
| test_precision | 0.9446428571428571 |
| test_recall | 0.9285714285714286 |
| test_f1 | 0.930292723149866 |
| test_runtime (s) | 4.1488 |
| test_samples_per_second | 6.749 |
| test_steps_per_second | 3.374 |
| epoch | 10.0 |
### Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0
- Datasets 2.10.1
- Tokenizers 0.13.2
|