Hugging Face's logo

yuweiiizz
/
whisper-small-taiwanese-lora

PEFT
TensorBoard
Safetensors
Min Nan Chinese
Generated from Trainer
Model card Files Metrics Community
yuweiiizz commited on
Commit
de0dcfe
•
1 Parent(s): 66f7ef7

End of training

Browse files
Files changed (3) hide show
  1. README.md +13 -11
  2. adapter_model.safetensors +1 -1
  3. runs/Apr19_15-21-09_ff068187d8b5/events.out.tfevents.1713540074.ff068187d8b5.26.0 +2 -2
README.md CHANGED
@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 16.1 dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.3913
24
 
25
  ## Model description
26
 
@@ -46,19 +46,21 @@ The following hyperparameters were used during training:
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 600
49
- - num_epochs: 6
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss |
55
- |:-------------:|:-----:|:----:|:---------------:|
56
- | 0.6127 | 1.0 | 1550 | 0.5854 |
57
- | 0.4298 | 2.0 | 3100 | 0.4626 |
58
- | 0.2312 | 3.0 | 4650 | 0.4104 |
59
- | 0.1738 | 4.0 | 6200 | 0.4112 |
60
- | 0.1321 | 5.0 | 7750 | 0.3960 |
61
- | 0.0718 | 6.0 | 9300 | 0.3913 |
 
 
62
 
63
 
64
  ### Framework versions
@@ -66,5 +68,5 @@ The following hyperparameters were used during training:
66
  - PEFT 0.10.1.dev0
67
  - Transformers 4.38.1
68
  - Pytorch 2.1.2
69
- - Datasets 2.18.1.dev0
70
  - Tokenizers 0.15.2
 
20
 
21
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 16.1 dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.3965
24
 
25
  ## Model description
26
 
 
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
  - lr_scheduler_warmup_steps: 600
49
+ - num_epochs: 8
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss |
55
+ |:-------------:|:-----:|:-----:|:---------------:|
56
+ | 0.6127 | 1.0 | 1550 | 0.5854 |
57
+ | 0.4298 | 2.0 | 3100 | 0.4626 |
58
+ | 0.2312 | 3.0 | 4650 | 0.4104 |
59
+ | 0.1738 | 4.0 | 6200 | 0.4112 |
60
+ | 0.1321 | 5.0 | 7750 | 0.3960 |
61
+ | 0.0898 | 6.0 | 9300 | 0.4087 |
62
+ | 0.0652 | 7.0 | 10850 | 0.3947 |
63
+ | 0.0341 | 8.0 | 12400 | 0.3965 |
64
 
65
 
66
  ### Framework versions
 
68
  - PEFT 0.10.1.dev0
69
  - Transformers 4.38.1
70
  - Pytorch 2.1.2
71
+ - Datasets 2.19.1.dev0
72
  - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:852ef1de8a6e7398fb75c9de92d164c7d22077814869109e912243eef642e3ef
3
  size 19548472
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4bfc02ed85a9c75775b9eab64008464e32d947e88793c21adebe1213a64d20c0
3
  size 19548472
runs/Apr19_15-21-09_ff068187d8b5/events.out.tfevents.1713540074.ff068187d8b5.26.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fdb1d265230b90e0fe3f7524f3cf6caacaa646e70f64cdca148e1fb93cfcbc99
3
- size 31309
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21ca9a69e6f57f49c075c20747273b0475b8d68c07c9155a0917867849d69c22
3
+ size 35310