Commit
•
200dff7
1
Parent(s):
18f97df
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- wer
|
7 |
+
model-index:
|
8 |
+
- name: wav2vec2-burak-new-v10-small
|
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-burak-new-v10-small
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.3345
|
20 |
+
- Wer: 0.2030
|
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: 16
|
41 |
+
- eval_batch_size: 8
|
42 |
+
- seed: 42
|
43 |
+
- gradient_accumulation_steps: 4
|
44 |
+
- total_train_batch_size: 64
|
45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
46 |
+
- lr_scheduler_type: linear
|
47 |
+
- lr_scheduler_warmup_steps: 500
|
48 |
+
- num_epochs: 271
|
49 |
+
- mixed_precision_training: Native AMP
|
50 |
+
|
51 |
+
### Training results
|
52 |
+
|
53 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
54 |
+
|:-------------:|:------:|:-----:|:---------------:|:------:|
|
55 |
+
| 6.1239 | 9.43 | 500 | 3.1263 | 1.0 |
|
56 |
+
| 1.7776 | 18.87 | 1000 | 0.3793 | 0.4838 |
|
57 |
+
| 0.5275 | 28.3 | 1500 | 0.2654 | 0.3379 |
|
58 |
+
| 0.3605 | 37.74 | 2000 | 0.2704 | 0.2953 |
|
59 |
+
| 0.2802 | 47.17 | 2500 | 0.2610 | 0.2911 |
|
60 |
+
| 0.2348 | 56.6 | 3000 | 0.2717 | 0.2677 |
|
61 |
+
| 0.2101 | 66.04 | 3500 | 0.2736 | 0.2691 |
|
62 |
+
| 0.1805 | 75.47 | 4000 | 0.2782 | 0.2595 |
|
63 |
+
| 0.1644 | 84.91 | 4500 | 0.2873 | 0.2491 |
|
64 |
+
| 0.1469 | 94.34 | 5000 | 0.3040 | 0.2381 |
|
65 |
+
| 0.138 | 103.77 | 5500 | 0.3205 | 0.2429 |
|
66 |
+
| 0.1247 | 113.21 | 6000 | 0.3217 | 0.2264 |
|
67 |
+
| 0.118 | 122.64 | 6500 | 0.3148 | 0.2244 |
|
68 |
+
| 0.1116 | 132.08 | 7000 | 0.3114 | 0.2209 |
|
69 |
+
| 0.1045 | 141.51 | 7500 | 0.3151 | 0.2175 |
|
70 |
+
| 0.0988 | 150.94 | 8000 | 0.3096 | 0.2092 |
|
71 |
+
| 0.0925 | 160.38 | 8500 | 0.3357 | 0.2230 |
|
72 |
+
| 0.0898 | 169.81 | 9000 | 0.3220 | 0.2099 |
|
73 |
+
| 0.0848 | 179.25 | 9500 | 0.3372 | 0.2209 |
|
74 |
+
| 0.0831 | 188.68 | 10000 | 0.3030 | 0.2030 |
|
75 |
+
| 0.0796 | 198.11 | 10500 | 0.3297 | 0.2127 |
|
76 |
+
| 0.0747 | 207.55 | 11000 | 0.3312 | 0.2134 |
|
77 |
+
| 0.0777 | 216.98 | 11500 | 0.3231 | 0.2168 |
|
78 |
+
| 0.0724 | 226.42 | 12000 | 0.3248 | 0.2078 |
|
79 |
+
| 0.0705 | 235.85 | 12500 | 0.3277 | 0.2023 |
|
80 |
+
| 0.0691 | 245.28 | 13000 | 0.3262 | 0.1996 |
|
81 |
+
| 0.0661 | 254.72 | 13500 | 0.3356 | 0.1996 |
|
82 |
+
| 0.0678 | 264.15 | 14000 | 0.3345 | 0.2030 |
|
83 |
+
|
84 |
+
|
85 |
+
### Framework versions
|
86 |
+
|
87 |
+
- Transformers 4.25.1
|
88 |
+
- Pytorch 1.13.0+cu116
|
89 |
+
- Datasets 2.7.1
|
90 |
+
- Tokenizers 0.13.2
|