EzraWilliam commited on
Commit
0897ebc
·
1 Parent(s): 5c0102a

End of training

Browse files
Files changed (1) hide show
  1. README.md +34 -14
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.5950598802395209
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the audiofolder dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.3026
36
- - Accuracy: 0.5951
37
 
38
  ## Model description
39
 
@@ -61,22 +61,42 @@ The following hyperparameters were used during training:
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_ratio: 0.1
64
- - num_epochs: 10
65
 
66
  ### Training results
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
- | 4.3203 | 0.98 | 31 | 3.2451 | 0.0487 |
71
- | 2.7385 | 2.0 | 63 | 2.5680 | 0.1175 |
72
- | 2.3513 | 2.98 | 94 | 2.2365 | 0.1557 |
73
- | 2.2073 | 4.0 | 126 | 2.0460 | 0.2156 |
74
- | 2.0125 | 4.98 | 157 | 1.8972 | 0.3361 |
75
- | 1.8194 | 6.0 | 189 | 1.6842 | 0.4124 |
76
- | 1.6046 | 6.98 | 220 | 1.5395 | 0.4648 |
77
- | 1.4933 | 8.0 | 252 | 1.4126 | 0.5644 |
78
- | 1.3973 | 8.98 | 283 | 1.3351 | 0.5636 |
79
- | 1.3323 | 9.84 | 310 | 1.3026 | 0.5951 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
80
 
81
 
82
  ### Framework versions
 
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.8181137724550899
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the audiofolder dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.7585
36
+ - Accuracy: 0.8181
37
 
38
  ## Model description
39
 
 
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_ratio: 0.1
64
+ - num_epochs: 30
65
 
66
  ### Training results
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | 8.0217 | 0.98 | 31 | 7.7416 | 0.0472 |
71
+ | 5.1076 | 2.0 | 63 | 3.5170 | 0.0472 |
72
+ | 3.0131 | 2.98 | 94 | 2.9921 | 0.0876 |
73
+ | 3.0119 | 4.0 | 126 | 2.9580 | 0.0928 |
74
+ | 2.685 | 4.98 | 157 | 2.6591 | 0.0793 |
75
+ | 2.4513 | 6.0 | 189 | 2.3831 | 0.1257 |
76
+ | 2.4415 | 6.98 | 220 | 2.3518 | 0.1415 |
77
+ | 2.2998 | 8.0 | 252 | 2.2327 | 0.1864 |
78
+ | 2.1987 | 8.98 | 283 | 2.1297 | 0.1549 |
79
+ | 2.1206 | 10.0 | 315 | 2.0529 | 0.2118 |
80
+ | 2.0542 | 10.98 | 346 | 1.9592 | 0.2507 |
81
+ | 1.9693 | 12.0 | 378 | 1.8652 | 0.2792 |
82
+ | 1.8677 | 12.98 | 409 | 1.7811 | 0.3668 |
83
+ | 1.7369 | 14.0 | 441 | 1.7902 | 0.2493 |
84
+ | 1.6551 | 14.98 | 472 | 1.6558 | 0.3406 |
85
+ | 1.6176 | 16.0 | 504 | 1.5724 | 0.3585 |
86
+ | 1.5666 | 16.98 | 535 | 1.5822 | 0.4207 |
87
+ | 1.5103 | 18.0 | 567 | 1.5028 | 0.4379 |
88
+ | 1.4695 | 18.98 | 598 | 1.4276 | 0.4970 |
89
+ | 1.3016 | 20.0 | 630 | 1.3621 | 0.4798 |
90
+ | 1.2025 | 20.98 | 661 | 1.2016 | 0.5778 |
91
+ | 1.1211 | 22.0 | 693 | 1.2346 | 0.5644 |
92
+ | 1.0204 | 22.98 | 724 | 1.0743 | 0.6445 |
93
+ | 0.9365 | 24.0 | 756 | 1.0121 | 0.6759 |
94
+ | 0.8553 | 24.98 | 787 | 0.9246 | 0.7290 |
95
+ | 0.7698 | 26.0 | 819 | 0.8603 | 0.7612 |
96
+ | 0.7336 | 26.98 | 850 | 0.8072 | 0.7867 |
97
+ | 0.6965 | 28.0 | 882 | 0.7770 | 0.8009 |
98
+ | 0.6662 | 28.98 | 913 | 0.7640 | 0.8136 |
99
+ | 0.63 | 29.52 | 930 | 0.7585 | 0.8181 |
100
 
101
 
102
  ### Framework versions