cristinakuo
commited on
Commit
•
9340e32
1
Parent(s):
0202d6a
update model card README.md
Browse files
README.md
CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
-
- Loss: 2.
|
18 |
- Wer: 1.0
|
19 |
|
20 |
## Model description
|
@@ -34,34 +34,32 @@ More information needed
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
-
- learning_rate: 0.
|
38 |
-
- train_batch_size:
|
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: 100
|
44 |
-
- num_epochs:
|
45 |
- mixed_precision_training: Native AMP
|
46 |
|
47 |
### Training results
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|:---:|
|
51 |
-
|
|
52 |
-
| 2.
|
53 |
-
| 2.
|
54 |
-
| 2.
|
55 |
-
| 2.
|
56 |
-
| 2.
|
57 |
-
| 2.
|
58 |
-
| 2.
|
59 |
-
| 2.
|
60 |
-
| 2.
|
61 |
-
| 2.
|
62 |
-
| 2.
|
63 |
-
| 2.8519 | 27.08 | 6500 | 2.8676 | 1.0 |
|
64 |
-
| 2.845 | 29.17 | 7000 | 2.8682 | 1.0 |
|
65 |
|
66 |
|
67 |
### Framework versions
|
|
|
14 |
|
15 |
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 2.8795
|
18 |
- Wer: 1.0
|
19 |
|
20 |
## Model description
|
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 0.0001
|
38 |
+
- train_batch_size: 32
|
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: 100
|
44 |
+
- num_epochs: 10
|
45 |
- mixed_precision_training: Native AMP
|
46 |
|
47 |
### Training results
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|:---:|
|
51 |
+
| 5.6846 | 0.83 | 100 | 2.9086 | 1.0 |
|
52 |
+
| 2.8686 | 1.67 | 200 | 2.8922 | 1.0 |
|
53 |
+
| 2.8805 | 2.5 | 300 | 2.9326 | 1.0 |
|
54 |
+
| 2.8613 | 3.33 | 400 | 2.8698 | 1.0 |
|
55 |
+
| 2.8643 | 4.17 | 500 | 2.9027 | 1.0 |
|
56 |
+
| 2.8688 | 5.0 | 600 | 2.9544 | 1.0 |
|
57 |
+
| 2.8689 | 5.83 | 700 | 2.8914 | 1.0 |
|
58 |
+
| 2.8558 | 6.67 | 800 | 2.8762 | 1.0 |
|
59 |
+
| 2.8537 | 7.5 | 900 | 2.8982 | 1.0 |
|
60 |
+
| 2.8522 | 8.33 | 1000 | 2.8820 | 1.0 |
|
61 |
+
| 2.8468 | 9.17 | 1100 | 2.8760 | 1.0 |
|
62 |
+
| 2.8454 | 10.0 | 1200 | 2.8795 | 1.0 |
|
|
|
|
|
63 |
|
64 |
|
65 |
### Framework versions
|