Hugging Face's logo

miosipof
/
whisper_medium_bbrave_12-11-2024_v3.1

PEFT
Safetensors
Italian
Generated from Trainer
Eval Results
Model card Files Community
miosipof commited on
Commit
83344ac
•
1 Parent(s): 7ddaeef

End of training

Browse files
Files changed (3) hide show
  1. README.md +8 -8
  2. adapter_model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  args: default
24
  metrics:
25
  - type: wer
26
- value: 61.71003717472119
27
  name: Wer
28
  ---
29
 
@@ -34,8 +34,8 @@ should probably proofread and complete it, then remove this comment. -->
34
 
35
  This model is a fine-tuned version of [b-brave/asr_double_training_15-10-2024_merged](https://huggingface.co/b-brave/asr_double_training_15-10-2024_merged) on the b-brave/speech_disorders_voice_edit dataset.
36
  It achieves the following results on the evaluation set:
37
- - Loss: 0.4531
38
- - Wer: 61.7100
39
 
40
  ## Model description
41
 
@@ -54,7 +54,7 @@ More information needed
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
- - learning_rate: 0.001
58
  - train_batch_size: 8
59
  - eval_batch_size: 4
60
  - seed: 42
@@ -62,7 +62,7 @@ The following hyperparameters were used during training:
62
  - total_train_batch_size: 16
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
- - lr_scheduler_warmup_steps: 100
66
  - num_epochs: 5
67
  - mixed_precision_training: Native AMP
68
 
@@ -70,9 +70,9 @@ The following hyperparameters were used during training:
70
 
71
  | Training Loss | Epoch | Step | Validation Loss | Wer |
72
  |:-------------:|:------:|:----:|:---------------:|:-------:|
73
- | 0.0956 | 1.3245 | 100 | 0.4690 | 39.2813 |
74
- | 0.0509 | 2.6490 | 200 | 0.5038 | 38.7856 |
75
- | 0.0189 | 3.9735 | 300 | 0.4531 | 61.7100 |
76
 
77
 
78
  ### Framework versions
 
23
  args: default
24
  metrics:
25
  - type: wer
26
+ value: 36.43122676579926
27
  name: Wer
28
  ---
29
 
 
34
 
35
  This model is a fine-tuned version of [b-brave/asr_double_training_15-10-2024_merged](https://huggingface.co/b-brave/asr_double_training_15-10-2024_merged) on the b-brave/speech_disorders_voice_edit dataset.
36
  It achieves the following results on the evaluation set:
37
+ - Loss: 0.4374
38
+ - Wer: 36.4312
39
 
40
  ## Model description
41
 
 
54
  ### Training hyperparameters
55
 
56
  The following hyperparameters were used during training:
57
+ - learning_rate: 0.0003
58
  - train_batch_size: 8
59
  - eval_batch_size: 4
60
  - seed: 42
 
62
  - total_train_batch_size: 16
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
+ - lr_scheduler_warmup_steps: 150
66
  - num_epochs: 5
67
  - mixed_precision_training: Native AMP
68
 
 
70
 
71
  | Training Loss | Epoch | Step | Validation Loss | Wer |
72
  |:-------------:|:------:|:----:|:---------------:|:-------:|
73
+ | 0.1082 | 1.3245 | 100 | 0.4437 | 36.4312 |
74
+ | 0.0678 | 2.6490 | 200 | 0.4386 | 36.3073 |
75
+ | 0.0364 | 3.9735 | 300 | 0.4374 | 36.4312 |
76
 
77
 
78
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bc6ebedb49dd68ba45abfdfb79f6578a76c8eab9228d4943aad6f3ddec7df3c7
3
  size 18915424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:baf3cbf7633db5dbd09afc0a7e0e5d122d354224de2ee153e9e4b82284c1e803
3
  size 18915424
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f9f416fe8c8813c22cbca41d9f8880a1e88c4c0a019e3dae688961ef3b5d805f
3
  size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:138e8e39685378aea814acd6584fbd5921836b061d78f42c8311cf44206191c5
3
  size 5368