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
fc7d9cf
·
verified ·
1 Parent(s): 967fb4f

End of training

Browse files
Files changed (3) hide show
  1. README.md +12 -14
  2. adapter_model.safetensors +1 -1
  3. runs/Apr05_13-10-14_5978cef41cb9/events.out.tfevents.1712322618.5978cef41cb9.26.0 +2 -2
README.md CHANGED
@@ -5,22 +5,22 @@ license: apache-2.0
5
  library_name: peft
6
  tags:
7
  - generated_from_trainer
 
8
  datasets:
9
  - mozilla-foundation/common_voice_16_1
10
- base_model: openai/whisper-small
11
  model-index:
12
- - name: Whisper Small Taiwanese - LoRA
13
  results: []
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
- # Whisper Small Taiwanese - LoRA
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: 0.6364
24
 
25
  ## Model description
26
 
@@ -41,28 +41,26 @@ More information needed
41
  The following hyperparameters were used during training:
42
  - learning_rate: 0.001
43
  - train_batch_size: 16
44
- - eval_batch_size: 24
45
  - seed: 42
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
- - lr_scheduler_warmup_steps: 250
49
- - num_epochs: 3
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss |
55
  |:-------------:|:-----:|:----:|:---------------:|
56
- | 0.9605 | 0.65 | 1000 | 0.9080 |
57
- | 0.7138 | 1.29 | 2000 | 0.7496 |
58
- | 0.6426 | 1.94 | 3000 | 0.6741 |
59
- | 0.5076 | 2.58 | 4000 | 0.6364 |
60
 
61
 
62
  ### Framework versions
63
 
64
- - PEFT 0.9.0
65
- - Transformers 4.38.2
66
  - Pytorch 2.1.2
67
- - Datasets 2.18.0
68
  - Tokenizers 0.15.2
 
5
  library_name: peft
6
  tags:
7
  - generated_from_trainer
8
+ base_model: openai/whisper-small
9
  datasets:
10
  - mozilla-foundation/common_voice_16_1
 
11
  model-index:
12
+ - name: Whisper small Taiwanese - LoRA
13
  results: []
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
+ # Whisper small Taiwanese - LoRA
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: 0.4459
24
 
25
  ## Model description
26
 
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 0.001
43
  - train_batch_size: 16
44
+ - eval_batch_size: 8
45
  - seed: 42
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
+ - lr_scheduler_warmup_steps: 50
49
+ - num_epochs: 2
50
  - mixed_precision_training: Native AMP
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss |
55
  |:-------------:|:-----:|:----:|:---------------:|
56
+ | 0.425 | 1.0 | 3111 | 0.5223 |
57
+ | 0.2723 | 2.0 | 6222 | 0.4459 |
 
 
58
 
59
 
60
  ### Framework versions
61
 
62
+ - PEFT 0.10.1.dev0
63
+ - Transformers 4.38.1
64
  - Pytorch 2.1.2
65
+ - Datasets 2.18.1.dev0
66
  - Tokenizers 0.15.2
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cdec3e820a30ed1ca55269a25c850835f724437b652d58aa10313c9d6b2d7bb6
3
  size 6564952
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12d11749b2bc70a450e363ed07b6a1ec88681e2a9b19ac755df4d7b10dc289db
3
  size 6564952
runs/Apr05_13-10-14_5978cef41cb9/events.out.tfevents.1712322618.5978cef41cb9.26.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ebf1c49145e6a281e961b49577d20ca3d2773a5074cf2c570a07b1aeece5fb6d
3
- size 56337
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56256c5e30c199e173bd41a3207a9d294308281cf143f2fd9b2d982b5eb91806
3
+ size 58650