Hugging Face's logo

dmusingu
/
muk-luganda-digits-classification

Audio Classification
Transformers
Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
Community
dmusingu commited on
Commit
1dd8de9
·
verified ·
1 Parent(s): 1b7d82b

End of training

Browse files
Files changed (3) hide show
  1. README.md +64 -54
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 3.4304
22
- - Accuracy: 0.4667
23
 
24
  ## Model description
25
 
@@ -38,69 +38,79 @@ More information needed
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
- - learning_rate: 3e-05
42
  - train_batch_size: 4
43
  - eval_batch_size: 1
44
  - seed: 42
45
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_ratio: 0.1
48
- - num_epochs: 50
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
- | 2.3048 | 1.0 | 60 | 2.2910 | 0.2333 |
55
- | 2.2968 | 2.0 | 120 | 2.2885 | 0.2333 |
56
- | 2.2833 | 3.0 | 180 | 2.2984 | 0.2 |
57
- | 2.2628 | 4.0 | 240 | 2.2951 | 0.1667 |
58
- | 2.2277 | 5.0 | 300 | 2.3184 | 0.1667 |
59
- | 2.1637 | 6.0 | 360 | 2.2936 | 0.1333 |
60
- | 2.0973 | 7.0 | 420 | 2.2544 | 0.1667 |
61
- | 1.9912 | 8.0 | 480 | 2.1642 | 0.2333 |
62
- | 1.8592 | 9.0 | 540 | 2.1926 | 0.3333 |
63
- | 1.7143 | 10.0 | 600 | 2.1636 | 0.3667 |
64
- | 1.6433 | 11.0 | 660 | 2.1392 | 0.3667 |
65
- | 1.4799 | 12.0 | 720 | 2.1999 | 0.3667 |
66
- | 1.3969 | 13.0 | 780 | 1.9897 | 0.3667 |
67
- | 1.2879 | 14.0 | 840 | 2.1673 | 0.4667 |
68
- | 1.2874 | 15.0 | 900 | 1.8996 | 0.4333 |
69
- | 1.0745 | 16.0 | 960 | 2.0203 | 0.4 |
70
- | 0.9496 | 17.0 | 1020 | 2.2862 | 0.3667 |
71
- | 0.8605 | 18.0 | 1080 | 2.0454 | 0.5 |
72
- | 0.7169 | 19.0 | 1140 | 2.0636 | 0.4 |
73
- | 0.6536 | 20.0 | 1200 | 2.3490 | 0.4 |
74
- | 0.6232 | 21.0 | 1260 | 2.3292 | 0.4 |
75
- | 0.5155 | 22.0 | 1320 | 2.3583 | 0.4 |
76
- | 0.4213 | 23.0 | 1380 | 2.2706 | 0.4667 |
77
- | 0.5417 | 24.0 | 1440 | 2.1699 | 0.5 |
78
- | 0.427 | 25.0 | 1500 | 2.6660 | 0.4333 |
79
- | 0.4919 | 26.0 | 1560 | 2.6427 | 0.4667 |
80
- | 0.3176 | 27.0 | 1620 | 2.7346 | 0.4333 |
81
- | 0.3232 | 28.0 | 1680 | 2.6299 | 0.5 |
82
- | 0.2715 | 29.0 | 1740 | 2.6801 | 0.5 |
83
- | 0.1921 | 30.0 | 1800 | 3.1908 | 0.4 |
84
- | 0.2325 | 31.0 | 1860 | 3.3189 | 0.4 |
85
- | 0.345 | 32.0 | 1920 | 2.8053 | 0.4667 |
86
- | 0.1883 | 33.0 | 1980 | 3.1182 | 0.4667 |
87
- | 0.2894 | 34.0 | 2040 | 3.4994 | 0.4333 |
88
- | 0.2821 | 35.0 | 2100 | 3.1150 | 0.4667 |
89
- | 0.2291 | 36.0 | 2160 | 2.9860 | 0.5 |
90
- | 0.2131 | 37.0 | 2220 | 3.2128 | 0.4667 |
91
- | 0.1971 | 38.0 | 2280 | 3.3228 | 0.4667 |
92
- | 0.2987 | 39.0 | 2340 | 3.4010 | 0.4667 |
93
- | 0.1688 | 40.0 | 2400 | 3.2529 | 0.4667 |
94
- | 0.219 | 41.0 | 2460 | 3.1388 | 0.4667 |
95
- | 0.137 | 42.0 | 2520 | 3.1523 | 0.5 |
96
- | 0.179 | 43.0 | 2580 | 3.3119 | 0.4667 |
97
- | 0.1648 | 44.0 | 2640 | 3.3644 | 0.4667 |
98
- | 0.1964 | 45.0 | 2700 | 3.4351 | 0.4667 |
99
- | 0.1602 | 46.0 | 2760 | 3.5707 | 0.4667 |
100
- | 0.1479 | 47.0 | 2820 | 3.5323 | 0.4667 |
101
- | 0.2026 | 48.0 | 2880 | 3.5060 | 0.4667 |
102
- | 0.1777 | 49.0 | 2940 | 3.4347 | 0.4667 |
103
- | 0.1391 | 50.0 | 3000 | 3.4304 | 0.4667 |
 
 
 
 
 
 
 
 
 
 
104
 
105
 
106
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 2.4964
22
+ - Accuracy: 0.3667
23
 
24
  ## Model description
25
 
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - learning_rate: 1e-05
42
  - train_batch_size: 4
43
  - eval_batch_size: 1
44
  - seed: 42
45
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_ratio: 0.1
48
+ - num_epochs: 60
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
53
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
54
+ | 2.3029 | 1.0 | 60 | 2.3031 | 0.0667 |
55
+ | 2.3018 | 2.0 | 120 | 2.3056 | 0.0667 |
56
+ | 2.2994 | 3.0 | 180 | 2.2999 | 0.1333 |
57
+ | 2.2935 | 4.0 | 240 | 2.2990 | 0.0667 |
58
+ | 2.287 | 5.0 | 300 | 2.3021 | 0.1 |
59
+ | 2.2777 | 6.0 | 360 | 2.3070 | 0.1 |
60
+ | 2.2558 | 7.0 | 420 | 2.3024 | 0.1 |
61
+ | 2.2326 | 8.0 | 480 | 2.3030 | 0.1 |
62
+ | 2.1915 | 9.0 | 540 | 2.2621 | 0.2 |
63
+ | 2.1423 | 10.0 | 600 | 2.2680 | 0.1667 |
64
+ | 2.079 | 11.0 | 660 | 2.2347 | 0.2333 |
65
+ | 2.0082 | 12.0 | 720 | 2.2083 | 0.2667 |
66
+ | 1.9388 | 13.0 | 780 | 2.1809 | 0.2667 |
67
+ | 1.8631 | 14.0 | 840 | 2.1715 | 0.3667 |
68
+ | 1.8218 | 15.0 | 900 | 2.1598 | 0.3 |
69
+ | 1.7112 | 16.0 | 960 | 2.1145 | 0.3333 |
70
+ | 1.6316 | 17.0 | 1020 | 2.1664 | 0.3 |
71
+ | 1.5869 | 18.0 | 1080 | 2.1852 | 0.3333 |
72
+ | 1.5166 | 19.0 | 1140 | 2.2421 | 0.3 |
73
+ | 1.4676 | 20.0 | 1200 | 2.2286 | 0.2667 |
74
+ | 1.4119 | 21.0 | 1260 | 2.0431 | 0.3333 |
75
+ | 1.355 | 22.0 | 1320 | 2.2048 | 0.3 |
76
+ | 1.2626 | 23.0 | 1380 | 2.0693 | 0.3 |
77
+ | 1.2237 | 24.0 | 1440 | 2.0888 | 0.3333 |
78
+ | 1.1686 | 25.0 | 1500 | 2.0231 | 0.4667 |
79
+ | 1.1277 | 26.0 | 1560 | 1.9774 | 0.4 |
80
+ | 1.1061 | 27.0 | 1620 | 1.9195 | 0.4 |
81
+ | 1.0296 | 28.0 | 1680 | 1.9851 | 0.4 |
82
+ | 0.9938 | 29.0 | 1740 | 1.9041 | 0.4 |
83
+ | 0.9104 | 30.0 | 1800 | 2.0739 | 0.4 |
84
+ | 0.8894 | 31.0 | 1860 | 2.0402 | 0.4333 |
85
+ | 0.8386 | 32.0 | 1920 | 2.0462 | 0.4333 |
86
+ | 0.7863 | 33.0 | 1980 | 2.0242 | 0.4 |
87
+ | 0.7464 | 34.0 | 2040 | 1.9713 | 0.4 |
88
+ | 0.7254 | 35.0 | 2100 | 2.0689 | 0.3667 |
89
+ | 0.6726 | 36.0 | 2160 | 2.1829 | 0.3333 |
90
+ | 0.6348 | 37.0 | 2220 | 2.1640 | 0.3333 |
91
+ | 0.6372 | 38.0 | 2280 | 2.1810 | 0.4 |
92
+ | 0.6318 | 39.0 | 2340 | 1.9867 | 0.4333 |
93
+ | 0.5468 | 40.0 | 2400 | 2.1383 | 0.4333 |
94
+ | 0.5434 | 41.0 | 2460 | 2.1016 | 0.3667 |
95
+ | 0.5148 | 42.0 | 2520 | 2.2827 | 0.3667 |
96
+ | 0.4843 | 43.0 | 2580 | 2.0657 | 0.4333 |
97
+ | 0.457 | 44.0 | 2640 | 2.1263 | 0.4 |
98
+ | 0.4493 | 45.0 | 2700 | 2.2703 | 0.3667 |
99
+ | 0.4155 | 46.0 | 2760 | 2.2780 | 0.4 |
100
+ | 0.4027 | 47.0 | 2820 | 2.2483 | 0.3667 |
101
+ | 0.4152 | 48.0 | 2880 | 2.2511 | 0.3667 |
102
+ | 0.3978 | 49.0 | 2940 | 2.3178 | 0.3333 |
103
+ | 0.3794 | 50.0 | 3000 | 2.3588 | 0.3667 |
104
+ | 0.3568 | 51.0 | 3060 | 2.4206 | 0.3667 |
105
+ | 0.3381 | 52.0 | 3120 | 2.4859 | 0.3333 |
106
+ | 0.3622 | 53.0 | 3180 | 2.4684 | 0.3667 |
107
+ | 0.376 | 54.0 | 3240 | 2.5034 | 0.3667 |
108
+ | 0.3348 | 55.0 | 3300 | 2.5368 | 0.3667 |
109
+ | 0.3438 | 56.0 | 3360 | 2.4634 | 0.4 |
110
+ | 0.3421 | 57.0 | 3420 | 2.4767 | 0.3667 |
111
+ | 0.3279 | 58.0 | 3480 | 2.4778 | 0.3333 |
112
+ | 0.337 | 59.0 | 3540 | 2.4917 | 0.3333 |
113
+ | 0.3212 | 60.0 | 3600 | 2.4964 | 0.3667 |
114
 
115
 
116
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a7fd29aa048af84926a41aef579c6ebc454adc0d2f1e6b4075ca73523b42448
3
  size 378310592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:787b5983b89d5f70eac78d6078e50a599208506602ceb6b1eafe839d9a569fd8
3
  size 378310592
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b043d570e56034a7c9dbb4d5fe6a1b2b5b113c30c30b36917ad98ba75c4ea87e
3
  size 5905
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16c5d7212b372916414694f85e82159484c3e33a5ad6fca6f45b4adad566d5ea
3
  size 5905