Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-3-old

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-3-old / README.md
fydhfzh's picture
fydhfzh
End of training
7d8948e verified
preview code
|
raw
history blame
10.4 kB
metadata
license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: hubert-classifier-aug-fold-3
    results: []

hubert-classifier-aug-fold-3

This model is a fine-tuned version of facebook/hubert-base-ls960 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5105
  • Accuracy: 0.8639
  • Precision: 0.8778
  • Recall: 0.8639
  • F1: 0.8624
  • Binary: 0.9049

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0001
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1 Binary
No log 0.22 50 3.8852 0.0445 0.0072 0.0445 0.0108 0.3195
No log 0.43 100 3.4756 0.0499 0.0263 0.0499 0.0140 0.3247
No log 0.65 150 3.2680 0.0916 0.0193 0.0916 0.0272 0.3611
No log 0.86 200 3.1133 0.1038 0.0266 0.1038 0.0369 0.3695
3.7026 1.08 250 2.9266 0.1765 0.0816 0.1765 0.0972 0.4212
3.7026 1.29 300 2.7455 0.2412 0.1540 0.2412 0.1516 0.4644
3.7026 1.51 350 2.4869 0.2898 0.1860 0.2898 0.1860 0.5008
3.7026 1.73 400 2.2598 0.3275 0.2377 0.3275 0.2426 0.5284
3.7026 1.94 450 2.0416 0.4232 0.3176 0.4232 0.3356 0.5947
2.8767 2.16 500 1.8498 0.4461 0.3751 0.4461 0.3752 0.6124
2.8767 2.37 550 1.6887 0.5202 0.4826 0.5202 0.4522 0.6639
2.8767 2.59 600 1.5765 0.5256 0.4880 0.5256 0.4630 0.6677
2.8767 2.8 650 1.4405 0.5930 0.5650 0.5930 0.5478 0.7152
2.2257 3.02 700 1.3663 0.6253 0.6037 0.6253 0.5834 0.7371
2.2257 3.24 750 1.2404 0.6577 0.6668 0.6577 0.6159 0.7602
2.2257 3.45 800 1.1946 0.6887 0.7128 0.6887 0.6654 0.7807
2.2257 3.67 850 1.0658 0.7251 0.7319 0.7251 0.7025 0.8073
2.2257 3.88 900 1.0567 0.7129 0.7529 0.7129 0.6874 0.7982
1.8296 4.1 950 1.0013 0.7372 0.7537 0.7372 0.7209 0.8151
1.8296 4.31 1000 0.9092 0.7534 0.7863 0.7534 0.7388 0.8284
1.8296 4.53 1050 0.8869 0.7574 0.7865 0.7574 0.7467 0.8313
1.8296 4.75 1100 0.8033 0.7763 0.8084 0.7763 0.7677 0.8446
1.8296 4.96 1150 0.7981 0.7790 0.8161 0.7790 0.7664 0.8451
1.6084 5.18 1200 0.7499 0.7978 0.8090 0.7978 0.7889 0.8588
1.6084 5.39 1250 0.7066 0.7938 0.8039 0.7938 0.7807 0.8550
1.6084 5.61 1300 0.7537 0.7938 0.8163 0.7938 0.7871 0.8555
1.6084 5.83 1350 0.7293 0.8113 0.8328 0.8113 0.8065 0.8687
1.4253 6.04 1400 0.7055 0.7992 0.8276 0.7992 0.7905 0.8593
1.4253 6.26 1450 0.6732 0.8059 0.8226 0.8059 0.8003 0.8648
1.4253 6.47 1500 0.6510 0.8315 0.8461 0.8315 0.8272 0.8838
1.4253 6.69 1550 0.6113 0.8329 0.8565 0.8329 0.8314 0.8814
1.4253 6.9 1600 0.6299 0.8248 0.8435 0.8248 0.8215 0.8787
1.2904 7.12 1650 0.6132 0.8410 0.8553 0.8410 0.8377 0.8900
1.2904 7.34 1700 0.5883 0.8464 0.8592 0.8464 0.8437 0.8938
1.2904 7.55 1750 0.5995 0.8437 0.8565 0.8437 0.8363 0.8908
1.2904 7.77 1800 0.5861 0.8450 0.8612 0.8450 0.8400 0.8923
1.2904 7.98 1850 0.6007 0.8410 0.8613 0.8410 0.8370 0.8889
1.2105 8.2 1900 0.5518 0.8437 0.8594 0.8437 0.8407 0.8904
1.2105 8.41 1950 0.5739 0.8342 0.8522 0.8342 0.8294 0.8846
1.2105 8.63 2000 0.5998 0.8275 0.8475 0.8275 0.8222 0.8806
1.2105 8.85 2050 0.5662 0.8531 0.8716 0.8531 0.8471 0.8978
1.1291 9.06 2100 0.5144 0.8639 0.8818 0.8639 0.8607 0.9040
1.1291 9.28 2150 0.4782 0.8625 0.8843 0.8625 0.8584 0.9036
1.1291 9.49 2200 0.4787 0.8612 0.8736 0.8612 0.8571 0.9026
1.1291 9.71 2250 0.4866 0.8666 0.8800 0.8666 0.8629 0.9050
1.1291 9.92 2300 0.5999 0.8342 0.8468 0.8342 0.8279 0.8846
1.0646 10.14 2350 0.5397 0.8518 0.8723 0.8518 0.8496 0.8962
1.0646 10.36 2400 0.4718 0.8585 0.8761 0.8585 0.8541 0.9005
1.0646 10.57 2450 0.4909 0.8625 0.8772 0.8625 0.8603 0.9035
1.0646 10.79 2500 0.4706 0.8544 0.8729 0.8544 0.8510 0.8988
1.0145 11.0 2550 0.4830 0.8558 0.8692 0.8558 0.8522 0.8993
1.0145 11.22 2600 0.4926 0.8652 0.8774 0.8652 0.8622 0.9055
1.0145 11.43 2650 0.4879 0.8693 0.8827 0.8693 0.8672 0.9086
1.0145 11.65 2700 0.5248 0.8666 0.8811 0.8666 0.8634 0.9063
1.0145 11.87 2750 0.4835 0.8585 0.8733 0.8585 0.8550 0.9012
0.9883 12.08 2800 0.4525 0.8814 0.8929 0.8814 0.8807 0.9177
0.9883 12.3 2850 0.4808 0.8706 0.8856 0.8706 0.8667 0.9101
0.9883 12.51 2900 0.4736 0.8720 0.8848 0.8720 0.8699 0.9111
0.9883 12.73 2950 0.4256 0.8774 0.8945 0.8774 0.8762 0.9146
0.9883 12.94 3000 0.4400 0.8841 0.8953 0.8841 0.8816 0.9182
0.9479 13.16 3050 0.4743 0.8666 0.8836 0.8666 0.8660 0.9067
0.9479 13.38 3100 0.4729 0.8760 0.8926 0.8760 0.8749 0.9133
0.9479 13.59 3150 0.4608 0.8733 0.8857 0.8733 0.8704 0.9105
0.9479 13.81 3200 0.4206 0.8868 0.8948 0.8868 0.8841 0.9216
0.8867 14.02 3250 0.4332 0.8881 0.8992 0.8881 0.8871 0.9201
0.8867 14.24 3300 0.4635 0.8733 0.8860 0.8733 0.8718 0.9108
0.8867 14.46 3350 0.4691 0.8827 0.8960 0.8827 0.8820 0.9170
0.8867 14.67 3400 0.4282 0.8774 0.8882 0.8774 0.8754 0.9140
0.8867 14.89 3450 0.4504 0.8801 0.8917 0.8801 0.8782 0.9162
0.8593 15.1 3500 0.4781 0.8760 0.8884 0.8760 0.8736 0.9125
0.8593 15.32 3550 0.4614 0.8895 0.8993 0.8895 0.8878 0.9230
0.8593 15.53 3600 0.4659 0.8774 0.8918 0.8774 0.8766 0.9151
0.8593 15.75 3650 0.4496 0.8814 0.8916 0.8814 0.8797 0.9164
0.8593 15.97 3700 0.4648 0.8827 0.8941 0.8827 0.8817 0.9186
0.8329 16.18 3750 0.4735 0.8827 0.8962 0.8827 0.8815 0.9182
0.8329 16.4 3800 0.4255 0.8935 0.9057 0.8935 0.8928 0.9252
0.8329 16.61 3850 0.4689 0.8747 0.8878 0.8747 0.8739 0.9117
0.8329 16.83 3900 0.4596 0.8841 0.8948 0.8841 0.8834 0.9183
0.8168 17.04 3950 0.4942 0.8760 0.8914 0.8760 0.8714 0.9123
0.8168 17.26 4000 0.5265 0.8747 0.8899 0.8747 0.8733 0.9117
0.8168 17.48 4050 0.4592 0.8787 0.8910 0.8787 0.8767 0.9150
0.8168 17.69 4100 0.4839 0.8693 0.8824 0.8693 0.8678 0.9080
0.8168 17.91 4150 0.4688 0.8827 0.8923 0.8827 0.8804 0.9173
0.7604 18.12 4200 0.4593 0.8733 0.8834 0.8733 0.8708 0.9098
0.7604 18.34 4250 0.4871 0.8760 0.8876 0.8760 0.8741 0.9117

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.3.0
  • Datasets 2.19.1
  • Tokenizers 0.15.1