Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-1

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-1 / README.md
fydhfzh's picture
fydhfzh
End of training
2ccf5b8 verified
preview code
|
raw
history blame contribute delete
10.8 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-1
    results: []

hubert-classifier-aug-fold-1

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: 0.6784
  • Accuracy: 0.8612
  • Precision: 0.8732
  • Recall: 0.8612
  • F1: 0.8586
  • Binary: 0.9035

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
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 100
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1 Binary
No log 0.24 50 4.4161 0.0180 0.0182 0.0180 0.0106 0.1504
No log 0.48 100 4.2802 0.0375 0.0091 0.0375 0.0074 0.3082
No log 0.72 150 3.9834 0.0502 0.0086 0.0502 0.0109 0.3226
No log 0.96 200 3.7286 0.0569 0.0236 0.0569 0.0159 0.3276
4.2271 1.2 250 3.4426 0.0891 0.0270 0.0891 0.0322 0.3596
4.2271 1.44 300 3.2540 0.1169 0.0601 0.1169 0.0604 0.3788
4.2271 1.68 350 3.1869 0.1176 0.0721 0.1176 0.0598 0.3683
4.2271 1.92 400 2.8711 0.1618 0.1145 0.1618 0.0984 0.4101
3.3668 2.16 450 2.6606 0.2644 0.1518 0.2644 0.1626 0.4816
3.3668 2.4 500 2.3190 0.3670 0.2659 0.3670 0.2721 0.5538
3.3668 2.63 550 2.0561 0.4120 0.3507 0.4120 0.3239 0.5857
3.3668 2.87 600 1.8485 0.4764 0.4155 0.4764 0.4052 0.6330
2.5092 3.11 650 1.7040 0.5296 0.4975 0.5296 0.4731 0.6697
2.5092 3.35 700 1.4804 0.5970 0.5614 0.5970 0.5443 0.7167
2.5092 3.59 750 1.3268 0.6434 0.6271 0.6434 0.6047 0.7488
2.5092 3.83 800 1.2244 0.6749 0.6423 0.6749 0.6342 0.7728
1.771 4.07 850 1.0787 0.7348 0.7587 0.7348 0.7168 0.8142
1.771 4.31 900 1.0527 0.7281 0.7380 0.7281 0.7070 0.8097
1.771 4.55 950 0.9342 0.7596 0.7759 0.7596 0.7454 0.8314
1.771 4.79 1000 0.8399 0.7880 0.7986 0.7880 0.7766 0.8507
1.3767 5.03 1050 0.8286 0.7970 0.8035 0.7970 0.7883 0.8575
1.3767 5.27 1100 0.8207 0.7888 0.8016 0.7888 0.7823 0.8524
1.3767 5.51 1150 0.7596 0.8112 0.8180 0.8112 0.8033 0.8690
1.3767 5.75 1200 0.7087 0.8067 0.8139 0.8067 0.8007 0.8658
1.3767 5.99 1250 0.7088 0.8045 0.8178 0.8045 0.7991 0.8637
1.1079 6.23 1300 0.7062 0.8150 0.8256 0.8150 0.8101 0.8698
1.1079 6.47 1350 0.6382 0.8285 0.8385 0.8285 0.8272 0.8810
1.1079 6.71 1400 0.6746 0.8240 0.8386 0.8240 0.8209 0.8783
1.1079 6.95 1450 0.6312 0.8367 0.8523 0.8367 0.8347 0.8867
0.9652 7.19 1500 0.6707 0.8255 0.8438 0.8255 0.8215 0.8775
0.9652 7.43 1550 0.6126 0.8479 0.8578 0.8479 0.8449 0.8942
0.9652 7.66 1600 0.6500 0.8427 0.8528 0.8427 0.8397 0.8912
0.9652 7.9 1650 0.6272 0.8412 0.8512 0.8412 0.8375 0.8885
0.8436 8.14 1700 0.6499 0.8509 0.8630 0.8509 0.8470 0.8970
0.8436 8.38 1750 0.6836 0.8337 0.8423 0.8337 0.8294 0.8841
0.8436 8.62 1800 0.6261 0.8487 0.8614 0.8487 0.8478 0.8951
0.8436 8.86 1850 0.5969 0.8584 0.8631 0.8584 0.8555 0.9019
0.7658 9.1 1900 0.6646 0.8397 0.8561 0.8397 0.8357 0.8872
0.7658 9.34 1950 0.5753 0.8644 0.8715 0.8644 0.8624 0.9049
0.7658 9.58 2000 0.6675 0.8404 0.8511 0.8404 0.8365 0.8885
0.7658 9.82 2050 0.6864 0.8360 0.8479 0.8360 0.8319 0.8859
0.6854 10.06 2100 0.6580 0.8479 0.8599 0.8479 0.8435 0.8948
0.6854 10.3 2150 0.6755 0.8509 0.8627 0.8509 0.8487 0.8963
0.6854 10.54 2200 0.6949 0.8524 0.8625 0.8524 0.8499 0.8969
0.6854 10.78 2250 0.7240 0.8434 0.8511 0.8434 0.8411 0.8905
0.6444 11.02 2300 0.6266 0.8502 0.8607 0.8502 0.8462 0.8950
0.6444 11.26 2350 0.6061 0.8674 0.8795 0.8674 0.8647 0.9073
0.6444 11.5 2400 0.6550 0.8509 0.8616 0.8509 0.8477 0.8955
0.6444 11.74 2450 0.6460 0.8457 0.8553 0.8457 0.8441 0.8913
0.6444 11.98 2500 0.5699 0.8577 0.8679 0.8577 0.8572 0.9010
0.6038 12.22 2550 0.6236 0.8517 0.8576 0.8517 0.8491 0.8963
0.6038 12.46 2600 0.5718 0.8674 0.8766 0.8674 0.8639 0.9071
0.6038 12.69 2650 0.5904 0.8644 0.8753 0.8644 0.8649 0.9061
0.6038 12.93 2700 0.6894 0.8487 0.8614 0.8487 0.8470 0.8951
0.5691 13.17 2750 0.6029 0.8652 0.8777 0.8652 0.8643 0.9064
0.5691 13.41 2800 0.6195 0.8727 0.8842 0.8727 0.8721 0.9105
0.5691 13.65 2850 0.6300 0.8682 0.8776 0.8682 0.8668 0.9076
0.5691 13.89 2900 0.6413 0.8644 0.8729 0.8644 0.8618 0.9058
0.5315 14.13 2950 0.7475 0.8509 0.8632 0.8509 0.8477 0.8958
0.5315 14.37 3000 0.6623 0.8659 0.8756 0.8659 0.8641 0.9069
0.5315 14.61 3050 0.6826 0.8547 0.8643 0.8547 0.8522 0.8978
0.5315 14.85 3100 0.6302 0.8712 0.8797 0.8712 0.8694 0.9097
0.5031 15.09 3150 0.5901 0.8787 0.8846 0.8787 0.8769 0.9157
0.5031 15.33 3200 0.6089 0.8652 0.8746 0.8652 0.8632 0.9056
0.5031 15.57 3250 0.6068 0.8719 0.8783 0.8719 0.8708 0.9108
0.5031 15.81 3300 0.6462 0.8652 0.8738 0.8652 0.8632 0.9056
0.4759 16.05 3350 0.6459 0.8607 0.8718 0.8607 0.8591 0.9013
0.4759 16.29 3400 0.6432 0.8644 0.8741 0.8644 0.8629 0.9052
0.4759 16.53 3450 0.6266 0.8652 0.8731 0.8652 0.8640 0.9058
0.4759 16.77 3500 0.5806 0.8824 0.8904 0.8824 0.8823 0.9170
0.4731 17.01 3550 0.6293 0.8697 0.8792 0.8697 0.8698 0.9089
0.4731 17.25 3600 0.6389 0.8682 0.8786 0.8682 0.8681 0.9079
0.4731 17.49 3650 0.6320 0.8712 0.8773 0.8712 0.8696 0.9098
0.4731 17.72 3700 0.6363 0.8742 0.8812 0.8742 0.8724 0.9128
0.4731 17.96 3750 0.6116 0.8854 0.8926 0.8854 0.8841 0.9199
0.4605 18.2 3800 0.6574 0.8794 0.8897 0.8794 0.8778 0.9161
0.4605 18.44 3850 0.6271 0.8749 0.8842 0.8749 0.8731 0.9135
0.4605 18.68 3900 0.6418 0.8749 0.8830 0.8749 0.8736 0.9139
0.4605 18.92 3950 0.6398 0.8704 0.8825 0.8704 0.8688 0.9103
0.4339 19.16 4000 0.6366 0.8689 0.8760 0.8689 0.8664 0.9085
0.4339 19.4 4050 0.6164 0.8727 0.8824 0.8727 0.8716 0.9110
0.4339 19.64 4100 0.6044 0.8846 0.8904 0.8846 0.8837 0.9190
0.4339 19.88 4150 0.6749 0.8742 0.8807 0.8742 0.8716 0.9123
0.4057 20.12 4200 0.7049 0.8637 0.8748 0.8637 0.8617 0.9059
0.4057 20.36 4250 0.6698 0.8727 0.8821 0.8727 0.8718 0.9116
0.4057 20.6 4300 0.6165 0.8779 0.8900 0.8779 0.8776 0.9146
0.4057 20.84 4350 0.5957 0.8697 0.8791 0.8697 0.8688 0.9087
0.4144 21.08 4400 0.6662 0.8644 0.8741 0.8644 0.8644 0.9047
0.4144 21.32 4450 0.7379 0.8487 0.8573 0.8487 0.8481 0.8942

Framework versions

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