Hugging Face's logo

Terps
/
distilhubert-finetuned-gtzan

Audio Classification
Transformers
PyTorch
hubert
Generated from Trainer
Model card Files Community
1
Terps commited on
Commit
db0b1d1
·
1 Parent(s): 4441426

End of training

Browse files
Files changed (2) hide show
  1. README.md +34 -34
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -4,7 +4,7 @@ base_model: ntu-spml/distilhubert
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
- - marsyas/gtzan-v2
8
  metrics:
9
  - accuracy
10
  model-index:
@@ -15,14 +15,14 @@ model-index:
15
  type: audio-classification
16
  dataset:
17
  name: GTZAN
18
- type: marsyas/gtzan-v2
19
  config: all
20
  split: train
21
  args: all
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.82
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 [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.7819
36
- - Accuracy: 0.82
37
 
38
  ## Model description
39
 
@@ -52,49 +52,49 @@ More information needed
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
- - learning_rate: 3e-05
56
  - train_batch_size: 8
57
  - eval_batch_size: 8
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
- - lr_scheduler_warmup_ratio: 0.3
62
  - num_epochs: 25
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
67
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
68
- | 2.286 | 1.0 | 113 | 2.2792 | 0.26 |
69
- | 2.1863 | 2.0 | 226 | 2.1408 | 0.34 |
70
- | 1.9386 | 3.0 | 339 | 1.8744 | 0.48 |
71
- | 1.6908 | 4.0 | 452 | 1.6502 | 0.57 |
72
- | 1.5259 | 5.0 | 565 | 1.4149 | 0.72 |
73
- | 1.1279 | 6.0 | 678 | 1.2700 | 0.62 |
74
- | 1.2204 | 7.0 | 791 | 0.9902 | 0.75 |
75
- | 0.861 | 8.0 | 904 | 0.8020 | 0.8 |
76
- | 0.8153 | 9.0 | 1017 | 0.7291 | 0.8 |
77
- | 0.3983 | 10.0 | 1130 | 0.7304 | 0.8 |
78
- | 0.2209 | 11.0 | 1243 | 0.6960 | 0.79 |
79
- | 0.2523 | 12.0 | 1356 | 0.5783 | 0.83 |
80
- | 0.1267 | 13.0 | 1469 | 0.5613 | 0.83 |
81
- | 0.0468 | 14.0 | 1582 | 0.7976 | 0.8 |
82
- | 0.025 | 15.0 | 1695 | 0.8478 | 0.81 |
83
- | 0.0158 | 16.0 | 1808 | 0.7448 | 0.8 |
84
- | 0.0706 | 17.0 | 1921 | 0.7183 | 0.83 |
85
- | 0.0096 | 18.0 | 2034 | 0.7532 | 0.82 |
86
- | 0.0076 | 19.0 | 2147 | 0.7907 | 0.81 |
87
- | 0.0354 | 20.0 | 2260 | 0.7120 | 0.83 |
88
- | 0.0063 | 21.0 | 2373 | 0.7525 | 0.83 |
89
- | 0.0055 | 22.0 | 2486 | 0.7647 | 0.82 |
90
- | 0.0049 | 23.0 | 2599 | 0.7945 | 0.82 |
91
- | 0.0048 | 24.0 | 2712 | 0.7982 | 0.82 |
92
- | 0.0321 | 25.0 | 2825 | 0.7819 | 0.82 |
93
 
94
 
95
  ### Framework versions
96
 
97
- - Transformers 4.34.0
98
- - Pytorch 2.0.1+cu118
99
  - Datasets 2.14.5
100
  - Tokenizers 0.14.1
 
4
  tags:
5
  - generated_from_trainer
6
  datasets:
7
+ - marsyas/gtzan
8
  metrics:
9
  - accuracy
10
  model-index:
 
15
  type: audio-classification
16
  dataset:
17
  name: GTZAN
18
+ type: marsyas/gtzan
19
  config: all
20
  split: train
21
  args: all
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.86
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 [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.5669
36
+ - Accuracy: 0.86
37
 
38
  ## Model description
39
 
 
52
  ### Training hyperparameters
53
 
54
  The following hyperparameters were used during training:
55
+ - learning_rate: 2e-05
56
  - train_batch_size: 8
57
  - eval_batch_size: 8
58
  - seed: 42
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
+ - lr_scheduler_warmup_ratio: 0.2
62
  - num_epochs: 25
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Accuracy |
67
  |:-------------:|:-----:|:----:|:---------------:|:--------:|
68
+ | 2.2857 | 1.0 | 113 | 2.2745 | 0.25 |
69
+ | 2.1795 | 2.0 | 226 | 2.1382 | 0.47 |
70
+ | 1.8958 | 3.0 | 339 | 1.8220 | 0.54 |
71
+ | 1.6475 | 4.0 | 452 | 1.5569 | 0.65 |
72
+ | 1.4246 | 5.0 | 565 | 1.3421 | 0.69 |
73
+ | 1.0504 | 6.0 | 678 | 1.1615 | 0.7 |
74
+ | 1.1759 | 7.0 | 791 | 1.0113 | 0.76 |
75
+ | 0.8636 | 8.0 | 904 | 0.8411 | 0.75 |
76
+ | 0.914 | 9.0 | 1017 | 0.7973 | 0.77 |
77
+ | 0.5748 | 10.0 | 1130 | 0.8049 | 0.79 |
78
+ | 0.4442 | 11.0 | 1243 | 0.7253 | 0.79 |
79
+ | 0.4276 | 12.0 | 1356 | 0.6600 | 0.8 |
80
+ | 0.3435 | 13.0 | 1469 | 0.5876 | 0.83 |
81
+ | 0.2779 | 14.0 | 1582 | 0.6596 | 0.82 |
82
+ | 0.2661 | 15.0 | 1695 | 0.5582 | 0.82 |
83
+ | 0.179 | 16.0 | 1808 | 0.5933 | 0.8 |
84
+ | 0.1559 | 17.0 | 1921 | 0.5518 | 0.8 |
85
+ | 0.1914 | 18.0 | 2034 | 0.5229 | 0.82 |
86
+ | 0.0899 | 19.0 | 2147 | 0.5910 | 0.85 |
87
+ | 0.2234 | 20.0 | 2260 | 0.5277 | 0.86 |
88
+ | 0.0578 | 21.0 | 2373 | 0.5493 | 0.84 |
89
+ | 0.0488 | 22.0 | 2486 | 0.5698 | 0.85 |
90
+ | 0.0322 | 23.0 | 2599 | 0.5713 | 0.86 |
91
+ | 0.0331 | 24.0 | 2712 | 0.5747 | 0.85 |
92
+ | 0.1019 | 25.0 | 2825 | 0.5669 | 0.86 |
93
 
94
 
95
  ### Framework versions
96
 
97
+ - Transformers 4.34.1
98
+ - Pytorch 2.1.0+cu118
99
  - Datasets 2.14.5
100
  - Tokenizers 0.14.1
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e78a4fb036af3ccd84ce17626840c21fd3b615236f1fdc3849138551625b3618
3
  size 94783885
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c75e20fb7d56f4b8a2742e035857ddffb92b82dac33013121f86e6648c96642e
3
  size 94783885