jiobiala24
commited on
Commit
•
3baa8d3
1
Parent(s):
b6b5b1a
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- common_voice
|
7 |
+
model-index:
|
8 |
+
- name: wav2vec2-base-checkpoint-11.1
|
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 |
+
# wav2vec2-base-checkpoint-11.1
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [jiobiala24/wav2vec2-base-checkpoint-10](https://huggingface.co/jiobiala24/wav2vec2-base-checkpoint-10) on the common_voice dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 1.0173
|
20 |
+
- Wer: 0.3350
|
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: 0.0001
|
40 |
+
- train_batch_size: 32
|
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: 1000
|
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 |
+
| 0.2788 | 1.52 | 1000 | 0.5776 | 0.3410 |
|
54 |
+
| 0.2277 | 3.04 | 2000 | 0.6148 | 0.3465 |
|
55 |
+
| 0.1772 | 4.56 | 3000 | 0.6497 | 0.3497 |
|
56 |
+
| 0.1528 | 6.08 | 4000 | 0.6786 | 0.3430 |
|
57 |
+
| 0.1285 | 7.6 | 5000 | 0.6779 | 0.3489 |
|
58 |
+
| 0.1104 | 9.12 | 6000 | 0.7417 | 0.3528 |
|
59 |
+
| 0.0965 | 10.64 | 7000 | 0.7956 | 0.3477 |
|
60 |
+
| 0.0914 | 12.16 | 8000 | 0.7994 | 0.3570 |
|
61 |
+
| 0.082 | 13.68 | 9000 | 0.8690 | 0.3510 |
|
62 |
+
| 0.0788 | 15.2 | 10000 | 0.8569 | 0.3526 |
|
63 |
+
| 0.0727 | 16.72 | 11000 | 0.8885 | 0.3440 |
|
64 |
+
| 0.0656 | 18.24 | 12000 | 0.9586 | 0.3476 |
|
65 |
+
| 0.0608 | 19.76 | 13000 | 0.9317 | 0.3495 |
|
66 |
+
| 0.0588 | 21.28 | 14000 | 0.9809 | 0.3449 |
|
67 |
+
| 0.0547 | 22.8 | 15000 | 0.9552 | 0.3421 |
|
68 |
+
| 0.0519 | 24.32 | 16000 | 0.9782 | 0.3380 |
|
69 |
+
| 0.0474 | 25.84 | 17000 | 0.9923 | 0.3386 |
|
70 |
+
| 0.046 | 27.36 | 18000 | 0.9984 | 0.3347 |
|
71 |
+
| 0.045 | 28.88 | 19000 | 1.0173 | 0.3350 |
|
72 |
+
|
73 |
+
|
74 |
+
### Framework versions
|
75 |
+
|
76 |
+
- Transformers 4.11.3
|
77 |
+
- Pytorch 1.10.0+cu111
|
78 |
+
- Datasets 1.13.3
|
79 |
+
- Tokenizers 0.10.3
|