patrickvonplaten commited on
Commit
bafcbc6
1 Parent(s): e0e92e6

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +100 -0
README.md ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - tr
4
+ tags:
5
+ - automatic-speech-recognition
6
+ - common_voice
7
+ - generated_from_trainer
8
+ datasets:
9
+ - common_voice
10
+ model-index:
11
+ - name: hello_2b_2
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # hello_2b_2
19
+
20
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-2b](https://huggingface.co/facebook/wav2vec2-xls-r-2b) on the COMMON_VOICE - TR dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.5324
23
+ - Wer: 0.5109
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 5e-06
43
+ - train_batch_size: 2
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 2
48
+ - gradient_accumulation_steps: 8
49
+ - total_train_batch_size: 32
50
+ - total_eval_batch_size: 16
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: linear
53
+ - lr_scheduler_warmup_steps: 500
54
+ - num_epochs: 30.0
55
+ - mixed_precision_training: Native AMP
56
+
57
+ ### Training results
58
+
59
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
60
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
61
+ | 3.3543 | 0.92 | 100 | 3.4342 | 1.0 |
62
+ | 3.0521 | 1.85 | 200 | 3.1243 | 1.0 |
63
+ | 1.4905 | 2.77 | 300 | 1.1760 | 0.9876 |
64
+ | 0.5852 | 3.7 | 400 | 0.7678 | 0.7405 |
65
+ | 0.4442 | 4.63 | 500 | 0.7637 | 0.7179 |
66
+ | 0.3816 | 5.55 | 600 | 0.7114 | 0.6726 |
67
+ | 0.2923 | 6.48 | 700 | 0.7109 | 0.6837 |
68
+ | 0.2771 | 7.4 | 800 | 0.6800 | 0.6530 |
69
+ | 0.1643 | 8.33 | 900 | 0.6031 | 0.6089 |
70
+ | 0.2931 | 9.26 | 1000 | 0.6467 | 0.6308 |
71
+ | 0.1495 | 10.18 | 1100 | 0.6042 | 0.6085 |
72
+ | 0.2093 | 11.11 | 1200 | 0.5850 | 0.5889 |
73
+ | 0.1329 | 12.04 | 1300 | 0.5557 | 0.5567 |
74
+ | 0.1005 | 12.96 | 1400 | 0.5964 | 0.5814 |
75
+ | 0.2162 | 13.88 | 1500 | 0.5692 | 0.5626 |
76
+ | 0.0923 | 14.81 | 1600 | 0.5508 | 0.5462 |
77
+ | 0.075 | 15.74 | 1700 | 0.5477 | 0.5307 |
78
+ | 0.2029 | 16.66 | 1800 | 0.5501 | 0.5300 |
79
+ | 0.0985 | 17.59 | 1900 | 0.5350 | 0.5303 |
80
+ | 0.1674 | 18.51 | 2000 | 0.5429 | 0.5241 |
81
+ | 0.1305 | 19.44 | 2100 | 0.5645 | 0.5443 |
82
+ | 0.0774 | 20.37 | 2200 | 0.5313 | 0.5216 |
83
+ | 0.1372 | 21.29 | 2300 | 0.5644 | 0.5392 |
84
+ | 0.1095 | 22.22 | 2400 | 0.5577 | 0.5306 |
85
+ | 0.0958 | 23.15 | 2500 | 0.5461 | 0.5273 |
86
+ | 0.0544 | 24.07 | 2600 | 0.5290 | 0.5055 |
87
+ | 0.0579 | 24.99 | 2700 | 0.5295 | 0.5150 |
88
+ | 0.1213 | 25.92 | 2800 | 0.5311 | 0.5221 |
89
+ | 0.0691 | 26.85 | 2900 | 0.5228 | 0.5095 |
90
+ | 0.1729 | 27.77 | 3000 | 0.5340 | 0.5095 |
91
+ | 0.0697 | 28.7 | 3100 | 0.5334 | 0.5139 |
92
+ | 0.0734 | 29.63 | 3200 | 0.5323 | 0.5140 |
93
+
94
+
95
+ ### Framework versions
96
+
97
+ - Transformers 4.13.0.dev0
98
+ - Pytorch 1.10.0
99
+ - Datasets 1.15.2.dev0
100
+ - Tokenizers 0.10.3