update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- wer
|
7 |
+
model-index:
|
8 |
+
- name: hubert-base-libri-demo-feature_extractor_not_frozen_v1_30epochs_weight_decay
|
9 |
+
results: []
|
10 |
+
---
|
11 |
+
|
12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
13 |
+
should probably proofread and complete it, then remove this comment. -->
|
14 |
+
|
15 |
+
# hubert-base-libri-demo-feature_extractor_not_frozen_v1_30epochs_weight_decay
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [facebook/hubert-base-ls960](https://huggingface.co/facebook/hubert-base-ls960) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 3.8346
|
20 |
+
- Wer: 1.0000
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 0.00015
|
40 |
+
- train_batch_size: 64
|
41 |
+
- eval_batch_size: 8
|
42 |
+
- seed: 42
|
43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
+
- lr_scheduler_type: linear
|
45 |
+
- lr_scheduler_warmup_steps: 3000
|
46 |
+
- num_epochs: 30
|
47 |
+
- mixed_precision_training: Native AMP
|
48 |
+
|
49 |
+
### Training results
|
50 |
+
|
51 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
52 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
53 |
+
| 4.3231 | 1.12 | 500 | 3.4739 | 1.0000 |
|
54 |
+
| 2.879 | 2.24 | 1000 | 3.5811 | 1.0000 |
|
55 |
+
| 2.8709 | 3.36 | 1500 | 3.9899 | 1.0000 |
|
56 |
+
| 2.871 | 4.48 | 2000 | 4.0805 | 1.0000 |
|
57 |
+
| 2.8706 | 5.61 | 2500 | 3.8464 | 1.0000 |
|
58 |
+
| 2.8672 | 6.73 | 3000 | 3.7938 | 1.0000 |
|
59 |
+
| 2.8709 | 7.85 | 3500 | 3.8578 | 1.0000 |
|
60 |
+
| 2.8708 | 8.97 | 4000 | 3.7691 | 1.0000 |
|
61 |
+
| 2.8678 | 10.09 | 4500 | 3.8619 | 1.0000 |
|
62 |
+
| 2.8662 | 11.21 | 5000 | 3.8804 | 1.0000 |
|
63 |
+
| 2.8664 | 12.33 | 5500 | 3.8169 | 1.0000 |
|
64 |
+
| 2.8662 | 13.45 | 6000 | 3.6758 | 1.0000 |
|
65 |
+
| 2.8654 | 14.57 | 6500 | 3.7314 | 1.0000 |
|
66 |
+
| 2.8658 | 15.7 | 7000 | 3.8113 | 1.0000 |
|
67 |
+
| 2.8647 | 16.82 | 7500 | 3.8938 | 1.0000 |
|
68 |
+
| 2.8653 | 17.94 | 8000 | 3.9268 | 1.0000 |
|
69 |
+
| 2.8652 | 19.06 | 8500 | 3.9288 | 1.0000 |
|
70 |
+
| 2.8649 | 20.18 | 9000 | 3.9164 | 1.0000 |
|
71 |
+
| 2.8654 | 21.3 | 9500 | 3.8781 | 1.0000 |
|
72 |
+
| 2.8652 | 22.42 | 10000 | 3.8628 | 1.0000 |
|
73 |
+
| 2.8642 | 23.54 | 10500 | 3.8646 | 1.0000 |
|
74 |
+
| 2.8655 | 24.66 | 11000 | 3.8467 | 1.0000 |
|
75 |
+
| 2.8658 | 25.78 | 11500 | 3.8157 | 1.0000 |
|
76 |
+
| 2.8647 | 26.91 | 12000 | 3.8274 | 1.0000 |
|
77 |
+
| 2.8648 | 28.03 | 12500 | 3.8022 | 1.0000 |
|
78 |
+
| 2.8644 | 29.15 | 13000 | 3.8346 | 1.0000 |
|
79 |
+
|
80 |
+
|
81 |
+
### Framework versions
|
82 |
+
|
83 |
+
- Transformers 4.30.0.dev0
|
84 |
+
- Pytorch 2.0.1
|
85 |
+
- Datasets 2.12.1.dev0
|
86 |
+
- Tokenizers 0.13.3
|