Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-2

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-2 / README.md
fydhfzh's picture
fydhfzh
End of training
d46ccbf verified
preview code
|
raw
history blame
12.6 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-2
    results: []

hubert-classifier-aug-fold-2

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: 0.7155
  • Accuracy: 0.8625
  • Precision: 0.8774
  • Recall: 0.8625
  • F1: 0.8616
  • 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
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 30
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1 Binary
No log 0.24 50 4.4213 0.0142 0.0068 0.0142 0.0043 0.1424
No log 0.48 100 4.3075 0.0442 0.0196 0.0442 0.0177 0.3070
No log 0.72 150 4.0138 0.0502 0.0284 0.0502 0.0150 0.3229
No log 0.96 200 3.7089 0.1049 0.0547 0.1049 0.0490 0.3679
4.2506 1.2 250 3.4627 0.1498 0.0995 0.1498 0.0882 0.3998
4.2506 1.44 300 3.2210 0.1753 0.0964 0.1753 0.0970 0.4209
4.2506 1.68 350 2.9579 0.2412 0.1495 0.2412 0.1554 0.4661
4.2506 1.92 400 2.6469 0.2869 0.2258 0.2869 0.1944 0.5003
3.3165 2.16 450 2.2886 0.4097 0.3315 0.4097 0.3156 0.5854
3.3165 2.4 500 2.0450 0.4757 0.4368 0.4757 0.4139 0.6314
3.3165 2.63 550 1.7683 0.5453 0.5445 0.5453 0.4891 0.6812
3.3165 2.87 600 1.6279 0.6090 0.6005 0.6090 0.5677 0.7214
2.2876 3.11 650 1.3926 0.6404 0.6494 0.6404 0.6015 0.7466
2.2876 3.35 700 1.2807 0.6689 0.6921 0.6689 0.6386 0.7679
2.2876 3.59 750 1.0819 0.7161 0.7222 0.7161 0.6956 0.8023
2.2876 3.83 800 1.0035 0.7326 0.7524 0.7326 0.7131 0.8132
1.6056 4.07 850 0.9128 0.7648 0.7692 0.7648 0.7466 0.8362
1.6056 4.31 900 0.8266 0.7910 0.7986 0.7910 0.7817 0.8542
1.6056 4.55 950 0.7800 0.7955 0.8165 0.7955 0.7839 0.8587
1.6056 4.79 1000 0.8136 0.7873 0.8078 0.7873 0.7747 0.8520
1.2478 5.03 1050 0.8057 0.7783 0.7988 0.7783 0.7707 0.8451
1.2478 5.27 1100 0.6884 0.8082 0.8287 0.8082 0.8051 0.8658
1.2478 5.51 1150 0.6515 0.8217 0.8312 0.8217 0.8173 0.8766
1.2478 5.75 1200 0.7116 0.8120 0.8295 0.8120 0.8047 0.8694
1.2478 5.99 1250 0.6343 0.8360 0.8492 0.8360 0.8337 0.8856
1.0252 6.23 1300 0.6608 0.8060 0.8252 0.8060 0.8055 0.8654
1.0252 6.47 1350 0.6766 0.8097 0.8265 0.8097 0.8068 0.8667
1.0252 6.71 1400 0.6744 0.8135 0.8298 0.8135 0.8108 0.8697
1.0252 6.95 1450 0.6163 0.8404 0.8539 0.8404 0.8391 0.8902
0.8918 7.19 1500 0.5671 0.8524 0.8627 0.8524 0.8517 0.8980
0.8918 7.43 1550 0.5416 0.8562 0.8689 0.8562 0.8546 0.9006
0.8918 7.66 1600 0.5710 0.8554 0.8648 0.8554 0.8534 0.8990
0.8918 7.9 1650 0.5341 0.8524 0.8649 0.8524 0.8508 0.8982
0.7857 8.14 1700 0.5764 0.8449 0.8554 0.8449 0.8429 0.8919
0.7857 8.38 1750 0.5792 0.8562 0.8646 0.8562 0.8544 0.8997
0.7857 8.62 1800 0.5455 0.8599 0.8689 0.8599 0.8566 0.9030
0.7857 8.86 1850 0.5837 0.8547 0.8655 0.8547 0.8521 0.8993
0.7152 9.1 1900 0.6607 0.8472 0.8592 0.8472 0.8459 0.8948
0.7152 9.34 1950 0.6106 0.8547 0.8648 0.8547 0.8538 0.8994
0.7152 9.58 2000 0.5548 0.8659 0.8762 0.8659 0.8648 0.9064
0.7152 9.82 2050 0.6189 0.8532 0.8635 0.8532 0.8514 0.8990
0.6356 10.06 2100 0.5869 0.8614 0.8718 0.8614 0.8601 0.9037
0.6356 10.3 2150 0.5585 0.8622 0.8707 0.8622 0.8603 0.9049
0.6356 10.54 2200 0.5857 0.8659 0.8781 0.8659 0.8642 0.9078
0.6356 10.78 2250 0.5226 0.8704 0.8806 0.8704 0.8699 0.9103
0.6081 11.02 2300 0.5549 0.8712 0.8849 0.8712 0.8703 0.9108
0.6081 11.26 2350 0.5454 0.8712 0.8829 0.8712 0.8707 0.9112
0.6081 11.5 2400 0.5374 0.8659 0.8737 0.8659 0.8648 0.9085
0.6081 11.74 2450 0.5789 0.8584 0.8742 0.8584 0.8580 0.9020
0.6081 11.98 2500 0.5650 0.8659 0.8779 0.8659 0.8653 0.9067
0.5506 12.22 2550 0.5908 0.8652 0.8750 0.8652 0.8651 0.9057
0.5506 12.46 2600 0.5970 0.8757 0.8809 0.8757 0.8749 0.9133
0.5506 12.69 2650 0.5703 0.8727 0.8811 0.8727 0.8726 0.9117
0.5506 12.93 2700 0.6146 0.8614 0.8716 0.8614 0.8607 0.9044
0.529 13.17 2750 0.5766 0.8742 0.8850 0.8742 0.8732 0.9124
0.529 13.41 2800 0.5620 0.8682 0.8778 0.8682 0.8670 0.9091
0.529 13.65 2850 0.5397 0.8846 0.8928 0.8846 0.8844 0.9211
0.529 13.89 2900 0.5858 0.8674 0.8779 0.8674 0.8671 0.9072
0.4938 14.13 2950 0.6406 0.8674 0.8770 0.8674 0.8669 0.9068
0.4938 14.37 3000 0.6754 0.8599 0.8696 0.8599 0.8583 0.9027
0.4938 14.61 3050 0.6439 0.8637 0.8734 0.8637 0.8623 0.9053
0.4938 14.85 3100 0.6580 0.8674 0.8781 0.8674 0.8661 0.9075
0.4645 15.09 3150 0.6317 0.8697 0.8783 0.8697 0.8690 0.9096
0.4645 15.33 3200 0.5905 0.8772 0.8865 0.8772 0.8764 0.9157
0.4645 15.57 3250 0.6268 0.8764 0.8847 0.8764 0.8751 0.9145
0.4645 15.81 3300 0.6298 0.8727 0.8808 0.8727 0.8719 0.9112
0.4467 16.05 3350 0.6039 0.8749 0.8828 0.8749 0.8736 0.9127
0.4467 16.29 3400 0.5955 0.8831 0.8890 0.8831 0.8823 0.9194
0.4467 16.53 3450 0.5954 0.8772 0.8865 0.8772 0.8761 0.9146
0.4467 16.77 3500 0.6088 0.8779 0.8857 0.8779 0.8773 0.9152
0.4269 17.01 3550 0.6572 0.8757 0.8830 0.8757 0.8748 0.9139
0.4269 17.25 3600 0.6490 0.8644 0.8730 0.8644 0.8626 0.9058
0.4269 17.49 3650 0.6591 0.8712 0.8794 0.8712 0.8696 0.9104
0.4269 17.72 3700 0.6369 0.8742 0.8854 0.8742 0.8726 0.9125
0.4269 17.96 3750 0.6000 0.8869 0.8922 0.8869 0.8863 0.9211
0.4099 18.2 3800 0.6395 0.8682 0.8763 0.8682 0.8665 0.9076
0.4099 18.44 3850 0.6416 0.8734 0.8795 0.8734 0.8727 0.9118
0.4099 18.68 3900 0.5822 0.8794 0.8855 0.8794 0.8785 0.9165
0.4099 18.92 3950 0.6365 0.8816 0.8885 0.8816 0.8806 0.9178
0.3945 19.16 4000 0.6389 0.8801 0.8864 0.8801 0.8791 0.9169
0.3945 19.4 4050 0.5901 0.8846 0.8908 0.8846 0.8843 0.9203
0.3945 19.64 4100 0.6197 0.8824 0.8879 0.8824 0.8816 0.9193
0.3945 19.88 4150 0.6365 0.8809 0.8875 0.8809 0.8798 0.9177
0.371 20.12 4200 0.6293 0.8831 0.8898 0.8831 0.8822 0.9190
0.371 20.36 4250 0.6753 0.8742 0.8815 0.8742 0.8732 0.9127
0.371 20.6 4300 0.6299 0.8757 0.8831 0.8757 0.8754 0.9140
0.371 20.84 4350 0.6034 0.8869 0.8950 0.8869 0.8863 0.9216
0.3772 21.08 4400 0.5806 0.8959 0.9025 0.8959 0.8955 0.9276
0.3772 21.32 4450 0.5994 0.8899 0.8952 0.8899 0.8890 0.9237
0.3772 21.56 4500 0.6378 0.8824 0.8895 0.8824 0.8819 0.9181
0.3772 21.8 4550 0.6267 0.8816 0.8869 0.8816 0.8808 0.9182
0.3512 22.04 4600 0.6104 0.8816 0.8878 0.8816 0.8804 0.9185
0.3512 22.28 4650 0.6146 0.8854 0.8907 0.8854 0.8845 0.9209
0.3512 22.51 4700 0.6408 0.8869 0.8944 0.8869 0.8866 0.9213
0.3512 22.75 4750 0.6246 0.8839 0.8906 0.8839 0.8833 0.9196
0.3512 22.99 4800 0.5967 0.8846 0.8911 0.8846 0.8837 0.9199
0.3339 23.23 4850 0.6727 0.8801 0.8867 0.8801 0.8795 0.9165
0.3339 23.47 4900 0.6122 0.8839 0.8905 0.8839 0.8833 0.9187
0.3339 23.71 4950 0.6191 0.8869 0.8932 0.8869 0.8860 0.9211
0.3339 23.95 5000 0.6397 0.8906 0.8974 0.8906 0.8898 0.9240
0.3172 24.19 5050 0.6519 0.8824 0.8905 0.8824 0.8821 0.9177
0.3172 24.43 5100 0.6141 0.8884 0.8948 0.8884 0.8877 0.9224
0.3172 24.67 5150 0.6530 0.8846 0.8912 0.8846 0.8836 0.9200
0.3172 24.91 5200 0.6152 0.8884 0.8938 0.8884 0.8875 0.9226
0.3049 25.15 5250 0.6550 0.8846 0.8920 0.8846 0.8833 0.9200
0.3049 25.39 5300 0.6510 0.8876 0.8938 0.8876 0.8867 0.9221
0.3049 25.63 5350 0.6402 0.8816 0.8883 0.8816 0.8812 0.9182

Framework versions

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