Hugging Face's logo

vonewman
/
wav2vec2-large-mms-1b-wolof

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Model card Files Community
vonewman commited on
Commit
302aefe
·
verified ·
1 Parent(s): cbcd81d

End of training

Browse files
Files changed (2) hide show
  1. README.md +51 -13
  2. adapter.wol.safetensors +1 -1
README.md CHANGED
@@ -1,9 +1,9 @@
1
  ---
2
- license: cc-by-nc-4.0
3
  library_name: transformers
 
 
4
  tags:
5
  - generated_from_trainer
6
- base_model: facebook/mms-1b-all
7
  metrics:
8
  - wer
9
  model-index:
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.3071
22
- - Wer: 0.3209
23
 
24
  ## Model description
25
 
@@ -38,13 +38,13 @@ More information needed
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
- - learning_rate: 0.001
42
- - train_batch_size: 64
43
- - eval_batch_size: 64
44
  - seed: 42
45
- - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
- - lr_scheduler_warmup_steps: 1000
48
  - num_epochs: 10
49
  - mixed_precision_training: Native AMP
50
 
@@ -52,12 +52,50 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer |
54
  |:-------------:|:------:|:----:|:---------------:|:------:|
55
- | 1.7202 | 5.0761 | 1000 | 0.3071 | 0.3209 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
56
 
57
 
58
  ### Framework versions
59
 
60
- - Transformers 4.46.0.dev0
61
- - Pytorch 2.1.0+cu118
62
- - Datasets 3.0.2
63
  - Tokenizers 0.20.1
 
1
  ---
 
2
  library_name: transformers
3
+ license: cc-by-nc-4.0
4
+ base_model: facebook/mms-1b-all
5
  tags:
6
  - generated_from_trainer
 
7
  metrics:
8
  - wer
9
  model-index:
 
18
 
19
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.3101
22
+ - Wer: 0.3232
23
 
24
  ## Model description
25
 
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - learning_rate: 0.0001
42
+ - train_batch_size: 32
43
+ - eval_batch_size: 32
44
  - seed: 42
45
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 100
48
  - num_epochs: 10
49
  - mixed_precision_training: Native AMP
50
 
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer |
54
  |:-------------:|:------:|:----:|:---------------:|:------:|
55
+ | 9.769 | 0.2538 | 100 | 3.0781 | 1.0000 |
56
+ | 1.2689 | 0.5076 | 200 | 0.4291 | 0.4298 |
57
+ | 0.5598 | 0.7614 | 300 | 0.3861 | 0.4024 |
58
+ | 0.54 | 1.0152 | 400 | 0.3721 | 0.3935 |
59
+ | 0.5216 | 1.2690 | 500 | 0.3615 | 0.3879 |
60
+ | 0.5151 | 1.5228 | 600 | 0.3552 | 0.3760 |
61
+ | 0.4979 | 1.7766 | 700 | 0.3498 | 0.3716 |
62
+ | 0.4911 | 2.0305 | 800 | 0.3456 | 0.3720 |
63
+ | 0.4879 | 2.2843 | 900 | 0.3428 | 0.3713 |
64
+ | 0.4951 | 2.5381 | 1000 | 0.3399 | 0.3690 |
65
+ | 0.4815 | 2.7919 | 1100 | 0.3373 | 0.3616 |
66
+ | 0.4728 | 3.0457 | 1200 | 0.3330 | 0.3548 |
67
+ | 0.4665 | 3.2995 | 1300 | 0.3321 | 0.3528 |
68
+ | 0.4744 | 3.5533 | 1400 | 0.3292 | 0.3474 |
69
+ | 0.4766 | 3.8071 | 1500 | 0.3268 | 0.3444 |
70
+ | 0.4666 | 4.0609 | 1600 | 0.3262 | 0.3456 |
71
+ | 0.4626 | 4.3147 | 1700 | 0.3241 | 0.3431 |
72
+ | 0.4683 | 4.5685 | 1800 | 0.3228 | 0.3376 |
73
+ | 0.459 | 4.8223 | 1900 | 0.3213 | 0.3351 |
74
+ | 0.461 | 5.0761 | 2000 | 0.3201 | 0.3357 |
75
+ | 0.4616 | 5.3299 | 2100 | 0.3192 | 0.3351 |
76
+ | 0.4561 | 5.5838 | 2200 | 0.3187 | 0.3310 |
77
+ | 0.4568 | 5.8376 | 2300 | 0.3178 | 0.3336 |
78
+ | 0.4523 | 6.0914 | 2400 | 0.3157 | 0.3308 |
79
+ | 0.4494 | 6.3452 | 2500 | 0.3151 | 0.3290 |
80
+ | 0.4485 | 6.5990 | 2600 | 0.3147 | 0.3297 |
81
+ | 0.4455 | 6.8528 | 2700 | 0.3143 | 0.3253 |
82
+ | 0.4574 | 7.1066 | 2800 | 0.3132 | 0.3247 |
83
+ | 0.4426 | 7.3604 | 2900 | 0.3128 | 0.3249 |
84
+ | 0.4478 | 7.6142 | 3000 | 0.3120 | 0.3242 |
85
+ | 0.4499 | 7.8680 | 3100 | 0.3116 | 0.3251 |
86
+ | 0.4517 | 8.1218 | 3200 | 0.3114 | 0.3223 |
87
+ | 0.4374 | 8.3756 | 3300 | 0.3108 | 0.3239 |
88
+ | 0.449 | 8.6294 | 3400 | 0.3104 | 0.3228 |
89
+ | 0.4444 | 8.8832 | 3500 | 0.3105 | 0.3232 |
90
+ | 0.4442 | 9.1371 | 3600 | 0.3102 | 0.3219 |
91
+ | 0.4463 | 9.3909 | 3700 | 0.3103 | 0.3228 |
92
+ | 0.4423 | 9.6447 | 3800 | 0.3100 | 0.3217 |
93
+ | 0.4382 | 9.8985 | 3900 | 0.3101 | 0.3232 |
94
 
95
 
96
  ### Framework versions
97
 
98
+ - Transformers 4.47.0.dev0
99
+ - Pytorch 2.3.0+cu121
100
+ - Datasets 2.18.0
101
  - Tokenizers 0.20.1
adapter.wol.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:77f6fe4eee654bdcae6c139358210c5905aabdb0d830ecee984600d8e5ed29ef
3
  size 8829276
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7759b6394a6b6d97cdf92ff696a33422c7062e55cc419ad17a55405a8c66904
3
  size 8829276