File size: 3,438 Bytes
fa6009b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
712a34f
9507816
 
 
 
 
fa6009b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
712a34f
fa6009b
 
 
 
 
712a34f
fa6009b
 
 
9507816
 
712a34f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fa6009b
 
 
 
 
 
 
 
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
---
base_model: ''
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: wav2GPT2MusiSD3200
  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. -->

# wav2GPT2MusiSD3200

This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8962
- Rouge1: 28.7212
- Rouge2: 7.4616
- Rougel: 21.8892
- Rougelsum: 21.8832
- Gen Len: 46.0

## 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-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Rouge1  | Rouge2 | Rougel  | Rougelsum | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
| 1.0361        | 1.0   | 1361  | 0.9081          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |
| 1.0015        | 2.0   | 2722  | 0.9021          | 28.1788 | 7.304  | 21.4695 | 21.4607   | 50.0    |
| 1.0003        | 3.0   | 4083  | 0.8976          | 27.7311 | 6.574  | 21.4759 | 21.4402   | 57.0    |
| 0.9761        | 4.0   | 5444  | 0.8914          | 27.7311 | 6.574  | 21.4759 | 21.4402   | 57.0    |
| 0.9928        | 5.0   | 6805  | 0.8884          | 27.7311 | 6.574  | 21.4759 | 21.4402   | 57.0    |
| 1.013         | 6.0   | 8166  | 0.8858          | 27.7311 | 6.574  | 21.4759 | 21.4402   | 57.0    |
| 1.0476        | 7.0   | 9527  | 0.8852          | 27.7311 | 6.574  | 21.4759 | 21.4402   | 57.0    |
| 1.0649        | 8.0   | 10888 | 0.8847          | 27.7311 | 6.574  | 21.4759 | 21.4402   | 57.0    |
| 1.1224        | 9.0   | 12249 | 0.8888          | 29.7298 | 7.2318 | 22.3598 | 22.3739   | 56.0    |
| 1.1818        | 10.0  | 13610 | 0.8949          | 29.7298 | 7.2318 | 22.3598 | 22.3739   | 56.0    |
| 1.1832        | 11.0  | 14971 | 0.8981          | 30.1982 | 7.2766 | 22.0853 | 22.118    | 59.0    |
| 1.1878        | 12.0  | 16332 | 0.8987          | 29.7298 | 7.2318 | 22.3598 | 22.3739   | 56.0    |
| 1.1833        | 13.0  | 17693 | 0.8983          | 29.7298 | 7.2318 | 22.3598 | 22.3739   | 56.0    |
| 1.1772        | 14.0  | 19054 | 0.8980          | 29.7298 | 7.2318 | 22.3598 | 22.3739   | 56.0    |
| 1.1723        | 15.0  | 20415 | 0.8974          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |
| 1.1778        | 16.0  | 21776 | 0.8972          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |
| 1.1707        | 17.0  | 23137 | 0.8968          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |
| 1.1767        | 18.0  | 24498 | 0.8964          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |
| 1.17          | 19.0  | 25859 | 0.8962          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |
| 1.1737        | 20.0  | 27220 | 0.8962          | 28.7212 | 7.4616 | 21.8892 | 21.8832   | 46.0    |


### Framework versions

- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.2
- Tokenizers 0.13.3