danieleV9H
commited on
Commit
•
c6f8f8c
1
Parent(s):
e43593b
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,92 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
model-index:
|
6 |
+
- name: hubert-base-timit-demo-google-colab-ft35ep
|
7 |
+
results: []
|
8 |
+
---
|
9 |
+
|
10 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
11 |
+
should probably proofread and complete it, then remove this comment. -->
|
12 |
+
|
13 |
+
# hubert-base-timit-demo-google-colab-ft35ep
|
14 |
+
|
15 |
+
This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on the None dataset.
|
16 |
+
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 0.4602
|
18 |
+
- Wer: 0.3466
|
19 |
+
|
20 |
+
## Model description
|
21 |
+
|
22 |
+
More information needed
|
23 |
+
|
24 |
+
## Intended uses & limitations
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Training and evaluation data
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training procedure
|
33 |
+
|
34 |
+
### Training hyperparameters
|
35 |
+
|
36 |
+
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 0.0001
|
38 |
+
- train_batch_size: 8
|
39 |
+
- eval_batch_size: 8
|
40 |
+
- seed: 42
|
41 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
+
- lr_scheduler_type: linear
|
43 |
+
- lr_scheduler_warmup_steps: 1000
|
44 |
+
- num_epochs: 30
|
45 |
+
- mixed_precision_training: Native AMP
|
46 |
+
|
47 |
+
### Training results
|
48 |
+
|
49 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
50 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
51 |
+
| 3.825 | 0.87 | 500 | 2.9521 | 1.0 |
|
52 |
+
| 2.431 | 1.73 | 1000 | 0.9760 | 0.8013 |
|
53 |
+
| 1.0089 | 2.6 | 1500 | 0.5934 | 0.5968 |
|
54 |
+
| 0.6859 | 3.46 | 2000 | 0.5132 | 0.5356 |
|
55 |
+
| 0.5302 | 4.33 | 2500 | 0.4506 | 0.4894 |
|
56 |
+
| 0.44 | 5.19 | 3000 | 0.4340 | 0.4670 |
|
57 |
+
| 0.3926 | 6.06 | 3500 | 0.4506 | 0.4528 |
|
58 |
+
| 0.3326 | 6.92 | 4000 | 0.4197 | 0.4486 |
|
59 |
+
| 0.2937 | 7.79 | 4500 | 0.4093 | 0.4193 |
|
60 |
+
| 0.2568 | 8.65 | 5000 | 0.4098 | 0.4229 |
|
61 |
+
| 0.2473 | 9.52 | 5500 | 0.4090 | 0.4141 |
|
62 |
+
| 0.2233 | 10.38 | 6000 | 0.4152 | 0.4125 |
|
63 |
+
| 0.2108 | 11.25 | 6500 | 0.4586 | 0.4189 |
|
64 |
+
| 0.2086 | 12.11 | 7000 | 0.4284 | 0.3969 |
|
65 |
+
| 0.1858 | 12.98 | 7500 | 0.4028 | 0.3946 |
|
66 |
+
| 0.1641 | 13.84 | 8000 | 0.4679 | 0.4002 |
|
67 |
+
| 0.1686 | 14.71 | 8500 | 0.4441 | 0.3936 |
|
68 |
+
| 0.1489 | 15.57 | 9000 | 0.4897 | 0.3828 |
|
69 |
+
| 0.1541 | 16.44 | 9500 | 0.4953 | 0.3783 |
|
70 |
+
| 0.1417 | 17.3 | 10000 | 0.4500 | 0.3758 |
|
71 |
+
| 0.1428 | 18.17 | 10500 | 0.4533 | 0.3796 |
|
72 |
+
| 0.1306 | 19.03 | 11000 | 0.4474 | 0.3792 |
|
73 |
+
| 0.1185 | 19.9 | 11500 | 0.4762 | 0.3743 |
|
74 |
+
| 0.1081 | 20.76 | 12000 | 0.4770 | 0.3699 |
|
75 |
+
| 0.1253 | 21.63 | 12500 | 0.4749 | 0.3629 |
|
76 |
+
| 0.1087 | 22.49 | 13000 | 0.4577 | 0.3534 |
|
77 |
+
| 0.1172 | 23.36 | 13500 | 0.4819 | 0.3525 |
|
78 |
+
| 0.1086 | 24.22 | 14000 | 0.4709 | 0.3623 |
|
79 |
+
| 0.089 | 25.09 | 14500 | 0.4852 | 0.3544 |
|
80 |
+
| 0.086 | 25.95 | 15000 | 0.4602 | 0.3555 |
|
81 |
+
| 0.086 | 26.82 | 15500 | 0.4861 | 0.3497 |
|
82 |
+
| 0.086 | 27.68 | 16000 | 0.4527 | 0.3473 |
|
83 |
+
| 0.0919 | 28.55 | 16500 | 0.4607 | 0.3487 |
|
84 |
+
| 0.0792 | 29.41 | 17000 | 0.4602 | 0.3466 |
|
85 |
+
|
86 |
+
|
87 |
+
### Framework versions
|
88 |
+
|
89 |
+
- Transformers 4.17.0
|
90 |
+
- Pytorch 1.11.0+cu113
|
91 |
+
- Datasets 1.18.3
|
92 |
+
- Tokenizers 0.12.1
|