Hugging Face's logo

thisisjibon
/
distilhubert-finetuned-banglabeats

Audio Classification
Transformers
PyTorch
hubert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
6
thisisjibon commited on
Commit
f727386
·
1 Parent(s): 4f5c835

End of training

Browse files
Files changed (1) hide show
  1. README.md +28 -16
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.81125
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 BanglaBeats dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.4221
36
- - Accuracy: 0.8113
37
 
38
  ## Model description
39
 
@@ -56,25 +56,37 @@ The following hyperparameters were used during training:
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: 0.1
62
- - num_epochs: 10
63
 
64
  ### Training results
65
 
66
- | Training Loss | Epoch | Step | Validation Loss | Accuracy |
67
- |:-------------:|:-----:|:----:|:---------------:|:--------:|
68
- | 0.8861 | 1.0 | 900 | 0.9921 | 0.6175 |
69
- | 0.7184 | 2.0 | 1800 | 0.8063 | 0.6913 |
70
- | 0.58 | 3.0 | 2700 | 0.6938 | 0.7562 |
71
- | 0.3803 | 4.0 | 3600 | 0.7527 | 0.7712 |
72
- | 0.172 | 5.0 | 4500 | 0.9628 | 0.77 |
73
- | 0.023 | 6.0 | 5400 | 1.2802 | 0.7863 |
74
- | 0.0004 | 7.0 | 6300 | 1.3272 | 0.8125 |
75
- | 0.0002 | 8.0 | 7200 | 1.4326 | 0.8037 |
76
- | 0.0005 | 9.0 | 8100 | 1.3734 | 0.8113 |
77
- | 0.0001 | 10.0 | 9000 | 1.4221 | 0.8113 |
 
 
 
 
 
 
 
 
 
 
78
 
79
 
80
  ### Framework versions
 
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
+ value: 0.8336425479282622
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 BanglaBeats dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 1.4126
36
+ - Accuracy: 0.8336
37
 
38
  ## Model description
39
 
 
56
  - train_batch_size: 8
57
  - eval_batch_size: 8
58
  - seed: 42
59
+ - gradient_accumulation_steps: 2
60
+ - total_train_batch_size: 16
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: 20
65
 
66
  ### Training results
67
 
68
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|
70
+ | 0.9439 | 1.0 | 910 | 0.9274 | 0.6425 |
71
+ | 0.854 | 2.0 | 1820 | 0.7498 | 0.7260 |
72
+ | 0.4835 | 3.0 | 2730 | 0.6329 | 0.7706 |
73
+ | 0.6226 | 4.0 | 3640 | 0.6159 | 0.7934 |
74
+ | 0.456 | 5.0 | 4550 | 0.7118 | 0.7972 |
75
+ | 0.0565 | 6.0 | 5460 | 0.7994 | 0.8052 |
76
+ | 0.2605 | 7.0 | 6370 | 0.9735 | 0.8151 |
77
+ | 0.3635 | 8.0 | 7280 | 1.0618 | 0.8244 |
78
+ | 0.1879 | 9.0 | 8190 | 1.1644 | 0.8213 |
79
+ | 0.0292 | 10.0 | 9100 | 1.2543 | 0.8194 |
80
+ | 0.0002 | 11.0 | 10010 | 1.4084 | 0.8101 |
81
+ | 0.0006 | 12.0 | 10920 | 1.3823 | 0.8132 |
82
+ | 0.088 | 13.0 | 11830 | 1.4016 | 0.8256 |
83
+ | 0.0381 | 14.0 | 12740 | 1.3587 | 0.8225 |
84
+ | 0.0 | 15.0 | 13650 | 1.4242 | 0.8169 |
85
+ | 0.0 | 16.0 | 14560 | 1.4053 | 0.8275 |
86
+ | 0.0183 | 17.0 | 15470 | 1.4357 | 0.8318 |
87
+ | 0.0 | 18.0 | 16380 | 1.4123 | 0.8306 |
88
+ | 0.0098 | 19.0 | 17290 | 1.4077 | 0.8330 |
89
+ | 0.0 | 20.0 | 18200 | 1.4126 | 0.8336 |
90
 
91
 
92
  ### Framework versions