update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
base_model: facebook/mms-1b-all
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- wer
|
8 |
+
model-index:
|
9 |
+
- name: wav2vec2-large-mms-1b-bemba-colab
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# wav2vec2-large-mms-1b-bemba-colab
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 0.1663
|
21 |
+
- Wer: 0.3303
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.001
|
41 |
+
- train_batch_size: 8
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- gradient_accumulation_steps: 2
|
45 |
+
- total_train_batch_size: 16
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: linear
|
48 |
+
- lr_scheduler_warmup_steps: 100
|
49 |
+
- num_epochs: 15
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
54 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
55 |
+
| 2.6643 | 0.26 | 200 | 0.2147 | 0.3900 |
|
56 |
+
| 0.4271 | 0.52 | 400 | 0.1996 | 0.3634 |
|
57 |
+
| 0.4004 | 0.77 | 600 | 0.1911 | 0.3595 |
|
58 |
+
| 0.3789 | 1.03 | 800 | 0.1905 | 0.3544 |
|
59 |
+
| 0.3707 | 1.29 | 1000 | 0.1821 | 0.3461 |
|
60 |
+
| 0.3811 | 1.55 | 1200 | 0.1815 | 0.3586 |
|
61 |
+
| 0.3662 | 1.8 | 1400 | 0.1811 | 0.3433 |
|
62 |
+
| 0.3627 | 2.06 | 1600 | 0.1814 | 0.3443 |
|
63 |
+
| 0.3529 | 2.32 | 1800 | 0.1807 | 0.3375 |
|
64 |
+
| 0.3466 | 2.58 | 2000 | 0.1758 | 0.3299 |
|
65 |
+
| 0.3481 | 2.84 | 2200 | 0.1781 | 0.3408 |
|
66 |
+
| 0.3446 | 3.09 | 2400 | 0.1761 | 0.3316 |
|
67 |
+
| 0.3379 | 3.35 | 2600 | 0.1702 | 0.3305 |
|
68 |
+
| 0.3371 | 3.61 | 2800 | 0.1668 | 0.3258 |
|
69 |
+
| 0.3326 | 3.87 | 3000 | 0.1661 | 0.3212 |
|
70 |
+
| 0.3297 | 4.12 | 3200 | 0.1706 | 0.3358 |
|
71 |
+
| 0.3267 | 4.38 | 3400 | 0.1707 | 0.3322 |
|
72 |
+
| 0.3328 | 4.64 | 3600 | 0.1663 | 0.3303 |
|
73 |
+
|
74 |
+
|
75 |
+
### Framework versions
|
76 |
+
|
77 |
+
- Transformers 4.31.0.dev0
|
78 |
+
- Pytorch 2.0.1+cu118
|
79 |
+
- Datasets 2.13.1
|
80 |
+
- Tokenizers 0.13.3
|