kingabzpro commited on
Commit
92d0074
1 Parent(s): 7b21f9a

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -4
README.md CHANGED
@@ -14,7 +14,11 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  # wav2vec2-large-xls-r-1b-Swedish
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 common_voice dataset.
 
 
 
 
18
 
19
  ## Model description
20
 
@@ -33,11 +37,11 @@ More information needed
33
  ### Training hyperparameters
34
 
35
  The following hyperparameters were used during training:
36
- - learning_rate: 0.0003
37
- - train_batch_size: 64
38
  - eval_batch_size: 8
39
  - seed: 42
40
- - gradient_accumulation_steps: 2
41
  - total_train_batch_size: 128
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: linear
@@ -45,6 +49,21 @@ The following hyperparameters were used during training:
45
  - num_epochs: 50
46
  - mixed_precision_training: Native AMP
47
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
  ### Framework versions
49
 
50
  - Transformers 4.17.0.dev0
 
14
 
15
  # wav2vec2-large-xls-r-1b-Swedish
16
 
17
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the common_voice dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.3370
20
+ - Wer: 0.1803
21
+ - Cer: 0.0569
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - learning_rate: 7.5e-05
41
+ - train_batch_size: 32
42
  - eval_batch_size: 8
43
  - seed: 42
44
+ - gradient_accumulation_steps: 4
45
  - total_train_batch_size: 128
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
 
49
  - num_epochs: 50
50
  - mixed_precision_training: Native AMP
51
 
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
55
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
56
+ | 3.1423 | 5.49 | 500 | 0.5523 | 0.4414 | 0.1313 |
57
+ | 0.8615 | 10.98 | 1000 | 0.3877 | 0.2946 | 0.0942 |
58
+ | 0.4848 | 16.48 | 1500 | 0.3580 | 0.2539 | 0.0798 |
59
+ | 0.3538 | 21.97 | 2000 | 0.3391 | 0.2254 | 0.0709 |
60
+ | 0.2879 | 27.47 | 2500 | 0.3392 | 0.2151 | 0.0680 |
61
+ | 0.2466 | 32.96 | 3000 | 0.3687 | 0.2131 | 0.0680 |
62
+ | 0.2146 | 38.46 | 3500 | 0.3551 | 0.1951 | 0.0618 |
63
+ | 0.1916 | 43.95 | 4000 | 0.3601 | 0.1867 | 0.0590 |
64
+ | 0.175 | 49.45 | 4500 | 0.3370 | 0.1803 | 0.0569 |
65
+
66
+
67
  ### Framework versions
68
 
69
  - Transformers 4.17.0.dev0