Hugging Face's logo

aydink
/
wav2vec2-base-finetuned-common_voice

Audio Classification
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
aydink commited on
Commit
f1c1dea
•
1 Parent(s): 0cefadc

Model save

Browse files
Files changed (2) hide show
  1. README.md +17 -17
  2. model.safetensors +1 -1
README.md CHANGED
@@ -20,13 +20,13 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.0870
24
- - Accuracy: 0.9875
25
- - F1: 0.9875
26
- - Recall: 0.9875
27
- - Precision: 0.9877
28
- - Mcc: 0.9844
29
- - Auc: 0.9968
30
 
31
  ## Model description
32
 
@@ -59,16 +59,16 @@ The following hyperparameters were used during training:
59
 
60
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision | Mcc | Auc |
61
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:---------:|:------:|:------:|
62
- | 0.9826 | 1.0 | 200 | 0.9330 | 0.715 | 0.6769 | 0.7150 | 0.7516 | 0.6708 | 0.9379 |
63
- | 0.2818 | 2.0 | 400 | 0.5294 | 0.8425 | 0.8362 | 0.8425 | 0.8731 | 0.8133 | 0.9738 |
64
- | 0.1229 | 3.0 | 600 | 0.2185 | 0.945 | 0.9455 | 0.945 | 0.9476 | 0.9317 | 0.9917 |
65
- | 0.0094 | 4.0 | 800 | 0.2905 | 0.9425 | 0.9428 | 0.9425 | 0.9476 | 0.9293 | 0.9932 |
66
- | 0.0256 | 5.0 | 1000 | 0.1565 | 0.97 | 0.9702 | 0.97 | 0.9720 | 0.9629 | 0.9972 |
67
- | 0.0032 | 6.0 | 1200 | 0.1577 | 0.9775 | 0.9775 | 0.9775 | 0.9778 | 0.9720 | 0.9941 |
68
- | 0.0869 | 7.0 | 1400 | 0.1017 | 0.9825 | 0.9824 | 0.9825 | 0.9826 | 0.9782 | 0.9965 |
69
- | 0.0019 | 8.0 | 1600 | 0.1194 | 0.9775 | 0.9776 | 0.9775 | 0.9783 | 0.9720 | 0.9967 |
70
- | 0.0017 | 9.0 | 1800 | 0.0947 | 0.985 | 0.9850 | 0.9850 | 0.9851 | 0.9813 | 0.9972 |
71
- | 0.0016 | 10.0 | 2000 | 0.0870 | 0.9875 | 0.9875 | 0.9875 | 0.9877 | 0.9844 | 0.9968 |
72
 
73
 
74
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.0560
24
+ - Accuracy: 0.99
25
+ - F1: 0.9900
26
+ - Recall: 0.99
27
+ - Precision: 0.9902
28
+ - Mcc: 0.9875
29
+ - Auc: 0.9983
30
 
31
  ## Model description
32
 
 
59
 
60
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision | Mcc | Auc |
61
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:---------:|:------:|:------:|
62
+ | 0.7394 | 1.0 | 200 | 0.2528 | 0.95 | 0.9497 | 0.95 | 0.9554 | 0.9390 | 0.9952 |
63
+ | 0.0013 | 2.0 | 400 | 0.0069 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 1.0 |
64
+ | 0.0435 | 3.0 | 600 | 0.0962 | 0.985 | 0.9850 | 0.9850 | 0.9852 | 0.9813 | 0.9987 |
65
+ | 0.1172 | 4.0 | 800 | 0.0434 | 0.995 | 0.9950 | 0.9950 | 0.9950 | 0.9938 | 0.9993 |
66
+ | 0.0005 | 5.0 | 1000 | 0.0496 | 0.9925 | 0.9925 | 0.9925 | 0.9926 | 0.9906 | 0.9984 |
67
+ | 0.0006 | 6.0 | 1200 | 0.0652 | 0.99 | 0.9900 | 0.99 | 0.9901 | 0.9875 | 0.9991 |
68
+ | 0.0004 | 7.0 | 1400 | 0.0267 | 0.9975 | 0.9975 | 0.9975 | 0.9975 | 0.9969 | 0.9982 |
69
+ | 0.0003 | 8.0 | 1600 | 0.0423 | 0.9925 | 0.9925 | 0.9925 | 0.9926 | 0.9906 | 0.9982 |
70
+ | 0.0003 | 9.0 | 1800 | 0.0549 | 0.9875 | 0.9875 | 0.9875 | 0.9877 | 0.9844 | 0.9982 |
71
+ | 0.0003 | 10.0 | 2000 | 0.0560 | 0.99 | 0.9900 | 0.99 | 0.9902 | 0.9875 | 0.9983 |
72
 
73
 
74
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46e0ad9c8ffbc09f06b17a3810800aefd62661958b0b266b8582d22125ac1cfc
3
  size 378305452
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc4ddb453d937b883fdb0220f480b5d63e0553246ae237541c678ed4dfc6acd8
3
  size 378305452