Hugging Face's logo

mouseyy
/
result_data-1

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Eval Results (legacy)
Model card Files
xet
Community
mouseyy commited on
Commit
ba8c43f
·
verified ·
1 Parent(s): d3f5dab

Model save

Browse files
Files changed (1) hide show
  1. README.md +16 -10
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Wer
25
  type: wer
26
- value: 1.0
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,9 +33,9 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  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_17_0 dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 17.1658
37
- - Wer: 1.0
38
- - Cer: 1.2793
39
 
40
  ## Model description
41
 
@@ -54,7 +54,7 @@ More information needed
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
- - learning_rate: 3.793766869035321e-05
58
  - train_batch_size: 16
59
  - eval_batch_size: 16
60
  - seed: 42
@@ -64,15 +64,21 @@ The following hyperparameters were used during training:
64
  - total_eval_batch_size: 32
65
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
66
  - lr_scheduler_type: linear
67
- - lr_scheduler_warmup_steps: 222
68
- - training_steps: 10
69
  - mixed_precision_training: Native AMP
70
 
71
  ### Training results
72
 
73
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
74
- |:-------------:|:------:|:----:|:---------------:|:---:|:------:|
75
- | No log | 0.0055 | 6 | 17.2325 | 1.0 | 1.3514 |
 
 
 
 
 
 
76
 
77
 
78
  ### Framework versions
 
23
  metrics:
24
  - name: Wer
25
  type: wer
26
+ value: 0.3637135578828191
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
33
 
34
  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_17_0 dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.2233
37
+ - Wer: 0.3637
38
+ - Cer: 0.1700
39
 
40
  ## Model description
41
 
 
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
+ - learning_rate: 6.532628754904162e-05
58
  - train_batch_size: 16
59
  - eval_batch_size: 16
60
  - seed: 42
 
64
  - total_eval_batch_size: 32
65
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
66
  - lr_scheduler_type: linear
67
+ - lr_scheduler_warmup_steps: 206
68
+ - num_epochs: 7.0
69
  - mixed_precision_training: Native AMP
70
 
71
  ### Training results
72
 
73
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
74
+ |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
75
+ | 0.6324 | 0.9099 | 1000 | 0.5004 | 0.6083 | 0.2381 |
76
+ | 0.3497 | 1.8198 | 2000 | 0.3087 | 0.4650 | 0.1965 |
77
+ | 0.2642 | 2.7298 | 3000 | 0.2636 | 0.4249 | 0.1841 |
78
+ | 0.2328 | 3.6397 | 4000 | 0.2431 | 0.3960 | 0.1789 |
79
+ | 0.1933 | 4.5496 | 5000 | 0.2289 | 0.3773 | 0.1732 |
80
+ | 0.1783 | 5.4595 | 6000 | 0.2300 | 0.3728 | 0.1711 |
81
+ | 0.1617 | 6.3694 | 7000 | 0.2233 | 0.3637 | 0.1700 |
82
 
83
 
84
  ### Framework versions