Eyvaz commited on
Commit
53e29aa
1 Parent(s): bb7a910

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +89 -0
README.md ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: wav2vec2-base-russian-demo-kaggle
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # wav2vec2-base-russian-demo-kaggle
14
+
15
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: inf
18
+ - Wer: 0.9997
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 0.0001
38
+ - train_batch_size: 12
39
+ - eval_batch_size: 8
40
+ - seed: 42
41
+ - gradient_accumulation_steps: 2
42
+ - total_train_batch_size: 24
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: linear
45
+ - lr_scheduler_warmup_steps: 1000
46
+ - num_epochs: 30
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
52
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
53
+ | 0.0102 | 1.03 | 500 | inf | 0.9997 |
54
+ | 0.0068 | 2.06 | 1000 | inf | 0.9997 |
55
+ | 0.0 | 3.09 | 1500 | inf | 0.9997 |
56
+ | 0.0313 | 4.12 | 2000 | inf | 0.9997 |
57
+ | 0.0 | 5.15 | 2500 | inf | 0.9997 |
58
+ | 0.0052 | 6.19 | 3000 | inf | 0.9997 |
59
+ | 0.0287 | 7.22 | 3500 | inf | 0.9997 |
60
+ | 0.0 | 8.25 | 4000 | inf | 0.9997 |
61
+ | 0.01 | 9.28 | 4500 | inf | 0.9997 |
62
+ | 0.0 | 10.31 | 5000 | inf | 0.9997 |
63
+ | 0.3919 | 11.34 | 5500 | inf | 0.9997 |
64
+ | 0.0 | 12.37 | 6000 | inf | 0.9997 |
65
+ | 0.0 | 13.4 | 6500 | inf | 0.9997 |
66
+ | 0.0 | 14.43 | 7000 | inf | 0.9997 |
67
+ | 0.6422 | 15.46 | 7500 | inf | 0.9997 |
68
+ | 0.0 | 16.49 | 8000 | inf | 0.9997 |
69
+ | 0.0 | 17.53 | 8500 | inf | 0.9997 |
70
+ | 0.0 | 18.56 | 9000 | inf | 0.9997 |
71
+ | 0.0 | 19.59 | 9500 | inf | 0.9997 |
72
+ | 0.0 | 20.62 | 10000 | inf | 0.9997 |
73
+ | 0.0427 | 21.65 | 10500 | inf | 0.9997 |
74
+ | 0.0 | 22.68 | 11000 | inf | 0.9997 |
75
+ | 0.0 | 23.71 | 11500 | inf | 0.9997 |
76
+ | 0.0 | 24.74 | 12000 | inf | 0.9997 |
77
+ | 0.0091 | 25.77 | 12500 | inf | 0.9997 |
78
+ | 0.1243 | 26.8 | 13000 | inf | 0.9997 |
79
+ | 0.0 | 27.83 | 13500 | inf | 0.9997 |
80
+ | 0.0 | 28.87 | 14000 | inf | 0.9997 |
81
+ | 0.0 | 29.9 | 14500 | inf | 0.9997 |
82
+
83
+
84
+ ### Framework versions
85
+
86
+ - Transformers 4.11.3
87
+ - Pytorch 1.9.1
88
+ - Datasets 1.13.3
89
+ - Tokenizers 0.10.3