Hugging Face's logo

hf-audio
/
w2v-bert-2.0-mongolian-colab-CV16.0

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2-bert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
ylacombe HF staff commited on
Commit
4d60b9b
•
1 Parent(s): 83682f4

End of training

Browse files
Files changed (3) hide show
  1. README.md +17 -11
  2. model.safetensors +1 -1
  3. runs/Jan11_15-48-45_c8593c749e03/events.out.tfevents.1704988226.c8593c749e03.18940.0 +2 -2
README.md CHANGED
@@ -21,7 +21,7 @@ model-index:
21
  metrics:
22
  - name: Wer
23
  type: wer
24
- value: 0.4041766369371329
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
31
 
32
  This model is a fine-tuned version of [ylacombe/w2v-bert-2.0](https://huggingface.co/ylacombe/w2v-bert-2.0) on the common_voice_16_0 dataset.
33
  It achieves the following results on the evaluation set:
34
- - Loss: 0.5160
35
- - Wer: 0.4042
36
 
37
  ## Model description
38
 
@@ -51,7 +51,7 @@ More information needed
51
  ### Training hyperparameters
52
 
53
  The following hyperparameters were used during training:
54
- - learning_rate: 8e-05
55
  - train_batch_size: 16
56
  - eval_batch_size: 8
57
  - seed: 42
@@ -60,19 +60,25 @@ The following hyperparameters were used during training:
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
  - lr_scheduler_warmup_steps: 500
63
- - num_epochs: 5
64
  - mixed_precision_training: Native AMP
65
 
66
  ### Training results
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Wer |
69
  |:-------------:|:-----:|:----:|:---------------:|:------:|
70
- | 3.6161 | 0.79 | 100 | 1.5583 | 0.9785 |
71
- | 0.7795 | 1.58 | 200 | 0.7328 | 0.5970 |
72
- | 0.5046 | 2.37 | 300 | 0.6393 | 0.5229 |
73
- | 0.4218 | 3.16 | 400 | 0.6404 | 0.5047 |
74
- | 0.3556 | 3.95 | 500 | 0.6419 | 0.4817 |
75
- | 0.2584 | 4.74 | 600 | 0.5160 | 0.4042 |
 
 
 
 
 
 
76
 
77
 
78
  ### Framework versions
 
21
  metrics:
22
  - name: Wer
23
  type: wer
24
+ value: 0.32330867957363496
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
 
32
  This model is a fine-tuned version of [ylacombe/w2v-bert-2.0](https://huggingface.co/ylacombe/w2v-bert-2.0) on the common_voice_16_0 dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: 0.5065
35
+ - Wer: 0.3233
36
 
37
  ## Model description
38
 
 
51
  ### Training hyperparameters
52
 
53
  The following hyperparameters were used during training:
54
+ - learning_rate: 5e-05
55
  - train_batch_size: 16
56
  - eval_batch_size: 8
57
  - seed: 42
 
60
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
  - lr_scheduler_type: linear
62
  - lr_scheduler_warmup_steps: 500
63
+ - num_epochs: 10
64
  - mixed_precision_training: Native AMP
65
 
66
  ### Training results
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Wer |
69
  |:-------------:|:-----:|:----:|:---------------:|:------:|
70
+ | 3.8092 | 0.79 | 100 | 2.1220 | 1.0404 |
71
+ | 0.9265 | 1.58 | 200 | 0.7650 | 0.6125 |
72
+ | 0.5241 | 2.37 | 300 | 0.6422 | 0.5244 |
73
+ | 0.4165 | 3.16 | 400 | 0.6275 | 0.4711 |
74
+ | 0.3393 | 3.95 | 500 | 0.6290 | 0.4884 |
75
+ | 0.2664 | 4.74 | 600 | 0.5784 | 0.4712 |
76
+ | 0.2315 | 5.53 | 700 | 0.5370 | 0.4160 |
77
+ | 0.1819 | 6.32 | 800 | 0.5268 | 0.3813 |
78
+ | 0.1339 | 7.11 | 900 | 0.5100 | 0.3643 |
79
+ | 0.0993 | 7.91 | 1000 | 0.5368 | 0.3549 |
80
+ | 0.0739 | 8.7 | 1100 | 0.5405 | 0.3378 |
81
+ | 0.055 | 9.49 | 1200 | 0.5065 | 0.3233 |
82
 
83
 
84
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5cf66cb9be96bd975cc91cfcd1ddb40638d2ea302d5a9f0b0c14ec655517bd6f
3
  size 2422974460
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5690caf9e157829637f40f28effd6418e3fa2591da69d1e69238b751522152a
3
  size 2422974460
runs/Jan11_15-48-45_c8593c749e03/events.out.tfevents.1704988226.c8593c749e03.18940.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9815c443f79bf22f0c129e40231147fd8c1a5f272c619aea50d5d5cd040c2be
3
- size 11341
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebf987a99f9fc037b98ae21a0bf71f00047eadc71f30ec09df415644e16f1de2
3
+ size 11695