Hugging Face's logo

Phakphum
/
outs

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
whisper
Generated from Trainer
Model card Files Metrics Community
Phakphum commited on
Commit
51d01e5
·
verified ·
1 Parent(s): 6bc1ee0

./outs

Browse files
Files changed (5) hide show
  1. README.md +22 -6
  2. model.safetensors +1 -1
  3. runs/Oct07_16-52-47_39c563fb672a/events.out.tfevents.1728319977.39c563fb672a.1117.0 +3 -0
  4. runs/Oct07_17-18-45_39c563fb672a/events.out.tfevents.1728321535.39c563fb672a.1117.1 +3 -0
  5. training_args.bin +2 -2
README.md CHANGED
@@ -15,6 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
15
  # outs
16
 
17
  This model is a fine-tuned version of [biodatlab/whisper-th-small-combined](https://huggingface.co/biodatlab/whisper-th-small-combined) on an unknown dataset.
 
 
 
18
 
19
  ## Model description
20
 
@@ -33,18 +36,31 @@ More information needed
33
  ### Training hyperparameters
34
 
35
  The following hyperparameters were used during training:
36
- - learning_rate: 8e-05
37
- - train_batch_size: 4
38
- - eval_batch_size: 4
39
  - seed: 42
 
 
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
- - num_epochs: 3.0
43
  - mixed_precision_training: Native AMP
44
 
 
 
 
 
 
 
 
 
 
 
 
45
  ### Framework versions
46
 
47
  - Transformers 4.44.2
48
- - Pytorch 2.4.0
49
- - Datasets 3.0.0
50
  - Tokenizers 0.19.1
 
15
  # outs
16
 
17
  This model is a fine-tuned version of [biodatlab/whisper-th-small-combined](https://huggingface.co/biodatlab/whisper-th-small-combined) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.1631
20
+ - Cer: 10.9836
21
 
22
  ## Model description
23
 
 
36
  ### Training hyperparameters
37
 
38
  The following hyperparameters were used during training:
39
+ - learning_rate: 5e-05
40
+ - train_batch_size: 16
41
+ - eval_batch_size: 16
42
  - seed: 42
43
+ - gradient_accumulation_steps: 2.0
44
+ - total_train_batch_size: 32.0
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
+ - num_epochs: 5.0
48
  - mixed_precision_training: Native AMP
49
 
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Cer |
53
+ |:-------------:|:------:|:----:|:---------------:|:-------:|
54
+ | No log | 0.9926 | 67 | 0.1876 | 10.2687 |
55
+ | 0.1249 | 2.0 | 135 | 0.1632 | 13.0653 |
56
+ | 0.0715 | 2.9926 | 202 | 0.1870 | 14.0862 |
57
+ | 0.0715 | 4.0 | 270 | 0.1636 | 10.3119 |
58
+ | 0.0386 | 4.9630 | 335 | 0.1631 | 10.9836 |
59
+
60
+
61
  ### Framework versions
62
 
63
  - Transformers 4.44.2
64
+ - Pytorch 2.4.1+cu121
65
+ - Datasets 3.0.1
66
  - Tokenizers 0.19.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8d15ebe7b4ab956b3cb0a42d74b2f0736117f0ad9908cc3c7684fc3d1d673a32
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b84f20058d18f3021a889f633d506d2d4383e4366bead3e1ec4ebac50d3b4db
3
  size 966995080
runs/Oct07_16-52-47_39c563fb672a/events.out.tfevents.1728319977.39c563fb672a.1117.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91be534fa4651ad7f011d9c775be0effce87658217448b259bc99f3e2edfb6f8
3
+ size 6522
runs/Oct07_17-18-45_39c563fb672a/events.out.tfevents.1728321535.39c563fb672a.1117.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6e556e4c148a5d329661aa6ac0e8b6f3d546e364b8aa92d6561831e640cc068
3
+ size 8253
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d457f9d613279f3d263fa3a54fc567d022ff4db02b8ce8271d6e49f33c5b065e
3
- size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a11774c0364ae268d6b5d7b0f960419f14db554a1462b7e2c38706e352ad4b4
3
+ size 5304