Hugging Face's logo

miosipof
/
asr_IT_AUG_Synth

PEFT
TensorBoard
Safetensors
Italian
Generated from Trainer
Eval Results
Model card Files Metrics Community
miosipof commited on
Commit
6e706cc
·
verified ·
1 Parent(s): a3cd18c

End of training

Browse files
Files changed (3) hide show
  1. README.md +20 -12
  2. adapter_model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -24,7 +24,7 @@ model-index:
24
  args: default
25
  metrics:
26
  - type: wer
27
- value: 141.63090128755366
28
  name: Wer
29
  ---
30
 
@@ -35,8 +35,8 @@ should probably proofread and complete it, then remove this comment. -->
35
 
36
  This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the ASR_Synthetic_Speecht5_TTS dataset.
37
  It achieves the following results on the evaluation set:
38
- - Loss: 9.8059
39
- - Wer: 141.6309
40
 
41
  ## Model description
42
 
@@ -55,7 +55,7 @@ More information needed
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
- - learning_rate: 0.001
59
  - train_batch_size: 4
60
  - eval_batch_size: 4
61
  - seed: 42
@@ -63,18 +63,26 @@ The following hyperparameters were used during training:
63
  - total_train_batch_size: 8
64
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
65
  - lr_scheduler_type: linear
66
- - lr_scheduler_warmup_steps: 50
67
- - training_steps: 100
68
  - mixed_precision_training: Native AMP
69
 
70
  ### Training results
71
 
72
- | Training Loss | Epoch | Step | Validation Loss | Wer |
73
- |:-------------:|:------:|:----:|:---------------:|:---------:|
74
- | 6.891 | 0.0244 | 25 | 4.4284 | 157.0815 |
75
- | 3.4032 | 0.0489 | 50 | 5.4864 | 1025.8941 |
76
- | 6.9892 | 0.0733 | 75 | 11.7361 | 120.7439 |
77
- | 7.1852 | 0.0978 | 100 | 9.8059 | 141.6309 |
 
 
 
 
 
 
 
 
78
 
79
 
80
  ### Framework versions
 
24
  args: default
25
  metrics:
26
  - type: wer
27
+ value: 75.9656652360515
28
  name: Wer
29
  ---
30
 
 
35
 
36
  This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the ASR_Synthetic_Speecht5_TTS dataset.
37
  It achieves the following results on the evaluation set:
38
+ - Loss: 3.1269
39
+ - Wer: 75.9657
40
 
41
  ## Model description
42
 
 
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
+ - learning_rate: 0.0003
59
  - train_batch_size: 4
60
  - eval_batch_size: 4
61
  - seed: 42
 
63
  - total_train_batch_size: 8
64
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
65
  - lr_scheduler_type: linear
66
+ - lr_scheduler_warmup_steps: 100
67
+ - training_steps: 300
68
  - mixed_precision_training: Native AMP
69
 
70
  ### Training results
71
 
72
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
73
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
74
+ | 7.8335 | 0.0244 | 25 | 7.0413 | 196.1373 |
75
+ | 6.5445 | 0.0489 | 50 | 5.3231 | 160.5150 |
76
+ | 4.2166 | 0.0733 | 75 | 3.6955 | 144.6352 |
77
+ | 2.2081 | 0.0978 | 100 | 3.5441 | 84.9785 |
78
+ | 1.815 | 0.1222 | 125 | 3.4335 | 83.4049 |
79
+ | 1.6465 | 0.1467 | 150 | 3.2851 | 136.9099 |
80
+ | 1.5241 | 0.1711 | 175 | 3.3021 | 357.3677 |
81
+ | 1.3811 | 0.1956 | 200 | 3.2476 | 81.5451 |
82
+ | 1.2553 | 0.2200 | 225 | 3.1495 | 132.1888 |
83
+ | 1.3158 | 0.2445 | 250 | 3.1816 | 76.9671 |
84
+ | 1.2942 | 0.2689 | 275 | 3.1349 | 74.3920 |
85
+ | 1.1935 | 0.2934 | 300 | 3.1269 | 75.9657 |
86
 
87
 
88
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7216e40357d0876847d51bfda57696a381fc1c69a520f6ccbf35a3f0d320afb3
3
  size 18915424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c112c2c6949c4a84873b4c2fa0d23f9f4d5083fe2fcc9dd568af36760b4bf937
3
  size 18915424
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:135a0844128cecbf7bee46c7644ac380f14fde27f9b6ed45411786bb64e3aaad
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79667d0c808e80408818bcea1c2414e7176125490e784e800e629fdfa2d763c9
3
  size 5304