File size: 1,830 Bytes
009c7c5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: bsd-3-clause
base_model: MIT/ast-finetuned-audioset-10-10-0.4593
tags:
- generated_from_trainer
metrics:
- accuracy
- wer
model-index:
- name: ast-finetuned-audioset-10-10-0.4593-keyword_spotting2
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# ast-finetuned-audioset-10-10-0.4593-keyword_spotting2

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 the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0123
- Accuracy: 0.8228
- Wer: 0.1772

## 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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | Wer    |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| 0.0008        | 1.0   | 28   | 1.3229          | 0.8228   | 0.1772 |
| 0.3356        | 2.0   | 56   | 1.3607          | 0.8228   | 0.1772 |
| 0.0035        | 3.0   | 84   | 1.0123          | 0.8228   | 0.1772 |
| 0.0051        | 4.0   | 112  | 1.5980          | 0.8228   | 0.1772 |
| 0.0001        | 5.0   | 140  | 1.4630          | 0.8228   | 0.1772 |


### Framework versions

- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.7.dev0
- Tokenizers 0.14.1