Val123val commited on
Commit
782c956
1 Parent(s): 5ed13d8

End of training

Browse files
README.md CHANGED
@@ -7,9 +7,22 @@ tags:
7
  - generated_from_trainer
8
  datasets:
9
  - bond005/sberdevices_golos_10h_crowd
 
 
10
  model-index:
11
  - name: my_model - Val123val
12
- results: []
 
 
 
 
 
 
 
 
 
 
 
13
  ---
14
 
15
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -18,6 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
18
  # my_model - Val123val
19
 
20
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Sberdevices_golos_10h_crowd dataset.
 
 
 
21
 
22
  ## Model description
23
 
@@ -37,17 +53,33 @@ More information needed
37
 
38
  The following hyperparameters were used during training:
39
  - learning_rate: 1e-05
40
- - train_batch_size: 64
41
- - eval_batch_size: 32
42
  - seed: 42
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
  - lr_scheduler_warmup_steps: 500
46
- - training_steps: 3000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
47
 
48
  ### Framework versions
49
 
50
  - Transformers 4.36.2
51
- - Pytorch 2.1.0+cu121
52
- - Datasets 2.16.1
53
  - Tokenizers 0.15.0
 
7
  - generated_from_trainer
8
  datasets:
9
  - bond005/sberdevices_golos_10h_crowd
10
+ metrics:
11
+ - wer
12
  model-index:
13
  - name: my_model - Val123val
14
+ results:
15
+ - task:
16
+ name: Automatic Speech Recognition
17
+ type: automatic-speech-recognition
18
+ dataset:
19
+ name: Sberdevices_golos_10h_crowd
20
+ type: bond005/sberdevices_golos_10h_crowd
21
+ args: 'split: test'
22
+ metrics:
23
+ - name: Wer
24
+ type: wer
25
+ value: 42.241139818232334
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
  # my_model - Val123val
32
 
33
  This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Sberdevices_golos_10h_crowd dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 0.1761
36
+ - Wer: 42.2411
37
 
38
  ## Model description
39
 
 
53
 
54
  The following hyperparameters were used during training:
55
  - learning_rate: 1e-05
56
+ - train_batch_size: 16
57
+ - eval_batch_size: 8
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
  - lr_scheduler_warmup_steps: 500
62
+ - training_steps: 5000
63
+
64
+ ### Training results
65
+
66
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
67
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|
68
+ | 0.1521 | 0.91 | 500 | 0.1824 | 29.3408 |
69
+ | 0.0824 | 1.82 | 1000 | 0.1702 | 27.5291 |
70
+ | 0.0304 | 2.73 | 1500 | 0.1726 | 45.1046 |
71
+ | 0.0114 | 3.64 | 2000 | 0.1704 | 40.1238 |
72
+ | 0.0039 | 4.55 | 2500 | 0.1692 | 32.1903 |
73
+ | 0.0013 | 5.45 | 3000 | 0.1704 | 34.0298 |
74
+ | 0.0029 | 6.36 | 3500 | 0.1712 | 39.8976 |
75
+ | 0.0007 | 7.27 | 4000 | 0.1738 | 39.4273 |
76
+ | 0.0006 | 8.18 | 4500 | 0.1755 | 41.0664 |
77
+ | 0.0005 | 9.09 | 5000 | 0.1761 | 42.2411 |
78
+
79
 
80
  ### Framework versions
81
 
82
  - Transformers 4.36.2
83
+ - Pytorch 2.1.2+cpu
84
+ - Datasets 2.16.0
85
  - Tokenizers 0.15.0
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:873da2b672f54054ef6f205d546acb1ae325b1b8a492391dfe39fca2f2b05318
3
  size 966995080
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a76eec8cccbb24c4bb14a94b17ff3be0dbc61d89fe0a3fe7e7665d0475cf7d1
3
  size 966995080
runs/Dec28_00-23-15_DESKTOP-1CUM2VF/events.out.tfevents.1703713101.DESKTOP-1CUM2VF.30008.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6e739f7ff78d0a0d76956f536a28f1f8762c5837bc57971a39678e6660861f83
3
- size 39760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8b05159059a5318a1fe548dca5eb34e73acf9e15919c51921859e0de3c4f95b
3
+ size 40114