Hugging Face's logo

ctaguchi
/
w2v-bert-2.0-gui

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2-bert
Generated from Trainer
Model card Files
xet
 Community
ctaguchi commited on
Commit
dfc07ca
·
verified ·
1 Parent(s): 094941e

Model save

Browse files
Files changed (2) hide show
  1. README.md +26 -26
  2. model.safetensors +1 -1
README.md CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 2.9724
20
- - Cer: 1.0
21
 
22
  ## Model description
23
 
@@ -44,7 +44,7 @@ The following hyperparameters were used during training:
44
  - total_train_batch_size: 16
45
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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
 
@@ -52,29 +52,29 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Cer |
54
  |:-------------:|:------:|:----:|:---------------:|:------:|
55
- | 15.1742 | 0.4329 | 100 | 3.0537 | 1.0 |
56
- | 5.9563 | 0.8658 | 200 | 2.9340 | 0.9839 |
57
- | 5.8952 | 1.2987 | 300 | 2.9200 | 0.9839 |
58
- | 5.8593 | 1.7316 | 400 | 2.9071 | 0.9839 |
59
- | 6.0686 | 2.1645 | 500 | 2.9044 | 0.9839 |
60
- | 5.8993 | 2.5974 | 600 | 2.8959 | 0.9839 |
61
- | 5.9040 | 3.0303 | 700 | 2.8811 | 0.9534 |
62
- | 5.8524 | 3.4632 | 800 | 2.8782 | 0.9564 |
63
- | 5.8113 | 3.8961 | 900 | 2.8453 | 0.9337 |
64
- | 5.7912 | 4.3290 | 1000 | 2.8231 | 0.9270 |
65
- | 5.7774 | 4.7619 | 1100 | 2.8235 | 0.9035 |
66
- | 5.8741 | 5.1948 | 1200 | 2.8550 | 0.9035 |
67
- | 5.8245 | 5.6277 | 1300 | 2.8957 | 0.9268 |
68
- | 6.2186 | 6.0606 | 1400 | 2.9482 | 1.0 |
69
- | 5.9860 | 6.4935 | 1500 | 2.9188 | 1.0 |
70
- | 5.9384 | 6.9264 | 1600 | 2.9292 | 1.0 |
71
- | 6.0080 | 7.3593 | 1700 | 2.9442 | 1.0 |
72
- | 6.0351 | 7.7922 | 1800 | 2.9402 | 1.0 |
73
- | 5.9574 | 8.2251 | 1900 | 2.9585 | 1.0 |
74
- | 6.0426 | 8.6580 | 2000 | 2.9658 | 1.0 |
75
- | 6.0568 | 9.0909 | 2100 | 2.9715 | 1.0 |
76
- | 6.0480 | 9.5238 | 2200 | 2.9722 | 1.0 |
77
- | 6.0058 | 9.9567 | 2300 | 2.9724 | 1.0 |
78
 
79
 
80
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 2.9872
20
+ - Cer: 0.9839
21
 
22
  ## Model description
23
 
 
44
  - total_train_batch_size: 16
45
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 500
48
  - num_epochs: 10
49
  - mixed_precision_training: Native AMP
50
 
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Cer |
54
  |:-------------:|:------:|:----:|:---------------:|:------:|
55
+ | 20.5172 | 0.4329 | 100 | 6.1927 | 1.0 |
56
+ | 9.1755 | 0.8658 | 200 | 3.0737 | 1.0 |
57
+ | 5.9769 | 1.2987 | 300 | 2.9275 | 0.9839 |
58
+ | 5.8767 | 1.7316 | 400 | 2.9258 | 0.9839 |
59
+ | 6.0773 | 2.1645 | 500 | 2.9087 | 0.9839 |
60
+ | 5.9225 | 2.5974 | 600 | 2.9052 | 0.9839 |
61
+ | 5.9328 | 3.0303 | 700 | 2.9009 | 0.9839 |
62
+ | 5.8895 | 3.4632 | 800 | 2.8969 | 0.9368 |
63
+ | 5.8888 | 3.8961 | 900 | 2.9218 | 0.9839 |
64
+ | 5.9117 | 4.3290 | 1000 | 2.9595 | 0.9672 |
65
+ | 5.9625 | 4.7619 | 1100 | 2.9033 | 0.9839 |
66
+ | 6.0566 | 5.1948 | 1200 | 2.9598 | 0.9839 |
67
+ | 5.9214 | 5.6277 | 1300 | 2.9107 | 0.9839 |
68
+ | 5.9976 | 6.0606 | 1400 | 2.9289 | 0.9839 |
69
+ | 5.9904 | 6.4935 | 1500 | 2.9166 | 0.9839 |
70
+ | 5.9354 | 6.9264 | 1600 | 2.9257 | 0.9839 |
71
+ | 6.0097 | 7.3593 | 1700 | 2.9428 | 0.9839 |
72
+ | 6.0392 | 7.7922 | 1800 | 2.9378 | 0.9839 |
73
+ | 5.9639 | 8.2251 | 1900 | 2.9657 | 0.9839 |
74
+ | 6.0595 | 8.6580 | 2000 | 2.9771 | 0.9839 |
75
+ | 6.0797 | 9.0909 | 2100 | 2.9865 | 0.9839 |
76
+ | 6.0741 | 9.5238 | 2200 | 2.9870 | 0.9839 |
77
+ | 6.0342 | 9.9567 | 2300 | 2.9872 | 0.9839 |
78
 
79
 
80
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2053e639e7694924e2f2ab55ea7833746526119065c1f14eff21f193bc75d50c
3
  size 2322234612
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58245c89b4b6f6b39e75940f867d1e4d82c4aff2dc345b2c79910bd463e9ad16
3
  size 2322234612