Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-4

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-4 / README.md
fydhfzh's picture
fydhfzh
End of training
5e74432 verified
preview code
|
raw
history blame
10.6 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-4
    results: []

hubert-classifier-aug-fold-4

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: 0.6864
  • Accuracy: 0.8612
  • Precision: 0.8735
  • Recall: 0.8612
  • F1: 0.8606
  • Binary: 0.9026

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 4.4183 0.0135 0.0052 0.0135 0.0050 0.1206
No log 0.48 100 4.2895 0.0412 0.0103 0.0412 0.0108 0.2549
No log 0.72 150 3.9961 0.0592 0.0278 0.0592 0.0286 0.3335
No log 0.96 200 3.7039 0.0742 0.0345 0.0742 0.0283 0.3461
4.2414 1.2 250 3.4082 0.1394 0.0770 0.1394 0.0791 0.3938
4.2414 1.44 300 3.1852 0.1979 0.1003 0.1979 0.1107 0.4377
4.2414 1.68 350 2.9114 0.2976 0.2089 0.2976 0.2077 0.5074
4.2414 1.92 400 2.5996 0.3366 0.2524 0.3366 0.2461 0.5348
3.2619 2.16 450 2.3085 0.4100 0.3088 0.4100 0.3208 0.5857
3.2619 2.4 500 2.0470 0.4873 0.4303 0.4873 0.4151 0.6394
3.2619 2.63 550 1.7979 0.5307 0.4934 0.5307 0.4713 0.6713
3.2619 2.87 600 1.5495 0.6012 0.5736 0.6012 0.5519 0.7199
2.2861 3.11 650 1.4334 0.6124 0.6399 0.6124 0.5690 0.7302
2.2861 3.35 700 1.3103 0.6544 0.6847 0.6544 0.6231 0.7576
2.2861 3.59 750 1.1635 0.6927 0.7090 0.6927 0.6698 0.7849
2.2861 3.83 800 1.0149 0.7549 0.7708 0.7549 0.7481 0.8289
1.5996 4.07 850 1.0088 0.7286 0.7489 0.7286 0.7069 0.8106
1.5996 4.31 900 0.8567 0.7699 0.7719 0.7699 0.7540 0.8398
1.5996 4.55 950 0.8231 0.7916 0.8071 0.7916 0.7830 0.8552
1.5996 4.79 1000 0.8333 0.7841 0.8041 0.7841 0.7754 0.8489
1.2436 5.03 1050 0.7799 0.7946 0.8084 0.7946 0.7884 0.8562
1.2436 5.27 1100 0.7429 0.7946 0.8097 0.7946 0.7917 0.8568
1.2436 5.51 1150 0.7050 0.8238 0.8384 0.8238 0.8202 0.8772
1.2436 5.75 1200 0.6639 0.8193 0.8251 0.8193 0.8139 0.8738
1.2436 5.99 1250 0.6918 0.8186 0.8387 0.8186 0.8141 0.8740
1.0055 6.23 1300 0.6296 0.8246 0.8383 0.8246 0.8186 0.8768
1.0055 6.47 1350 0.6581 0.8253 0.8412 0.8253 0.8215 0.8779
1.0055 6.71 1400 0.6302 0.8313 0.8469 0.8313 0.8292 0.8821
1.0055 6.95 1450 0.5998 0.8448 0.8593 0.8448 0.8420 0.8906
0.8805 7.19 1500 0.6230 0.8456 0.8531 0.8456 0.8443 0.8926
0.8805 7.43 1550 0.6308 0.8396 0.8528 0.8396 0.8377 0.8887
0.8805 7.66 1600 0.5979 0.8418 0.8531 0.8418 0.8413 0.8897
0.8805 7.9 1650 0.6022 0.8478 0.8616 0.8478 0.8469 0.8946
0.7681 8.14 1700 0.5872 0.8471 0.8573 0.8471 0.8453 0.8927
0.7681 8.38 1750 0.5744 0.8433 0.8570 0.8433 0.8420 0.8909
0.7681 8.62 1800 0.5351 0.8643 0.8707 0.8643 0.8627 0.9061
0.7681 8.86 1850 0.5688 0.8561 0.8680 0.8561 0.8559 0.8990
0.7001 9.1 1900 0.6618 0.8298 0.8457 0.8298 0.8269 0.8810
0.7001 9.34 1950 0.6244 0.8426 0.8571 0.8426 0.8422 0.8900
0.7001 9.58 2000 0.5802 0.8576 0.8681 0.8576 0.8569 0.8996
0.7001 9.82 2050 0.5352 0.8688 0.8761 0.8688 0.8687 0.9072
0.6288 10.06 2100 0.5347 0.8651 0.8773 0.8651 0.8637 0.9049
0.6288 10.3 2150 0.6019 0.8546 0.8665 0.8546 0.8535 0.8986
0.6288 10.54 2200 0.5699 0.8598 0.8670 0.8598 0.8571 0.9005
0.6288 10.78 2250 0.5494 0.8748 0.8838 0.8748 0.8730 0.9118
0.5959 11.02 2300 0.5471 0.8718 0.8804 0.8718 0.8714 0.9103
0.5959 11.26 2350 0.5570 0.8628 0.8738 0.8628 0.8605 0.9042
0.5959 11.5 2400 0.5300 0.8801 0.8875 0.8801 0.8791 0.9163
0.5959 11.74 2450 0.5418 0.8643 0.8725 0.8643 0.8630 0.9039
0.5959 11.98 2500 0.5418 0.8726 0.8822 0.8726 0.8715 0.9108
0.5407 12.22 2550 0.5718 0.8658 0.8755 0.8658 0.8652 0.9058
0.5407 12.46 2600 0.5686 0.8658 0.8725 0.8658 0.8643 0.9058
0.5407 12.69 2650 0.6045 0.8658 0.8768 0.8658 0.8656 0.9053
0.5407 12.93 2700 0.5571 0.8621 0.8715 0.8621 0.8607 0.9027
0.5175 13.17 2750 0.5367 0.8756 0.8809 0.8756 0.8745 0.9131
0.5175 13.41 2800 0.5241 0.8771 0.8827 0.8771 0.8755 0.9142
0.5175 13.65 2850 0.5793 0.8703 0.8792 0.8703 0.8691 0.9095
0.5175 13.89 2900 0.5608 0.8756 0.8843 0.8756 0.8751 0.9123
0.4913 14.13 2950 0.5734 0.8711 0.8781 0.8711 0.8694 0.9100
0.4913 14.37 3000 0.5916 0.8771 0.8821 0.8771 0.8758 0.9134
0.4913 14.61 3050 0.5651 0.8696 0.8761 0.8696 0.8680 0.9082
0.4913 14.85 3100 0.5535 0.8786 0.8831 0.8786 0.8771 0.9152
0.4747 15.09 3150 0.5694 0.8741 0.8819 0.8741 0.8737 0.9118
0.4747 15.33 3200 0.5759 0.8726 0.8794 0.8726 0.8720 0.9103
0.4747 15.57 3250 0.5827 0.8666 0.8718 0.8666 0.8642 0.9070
0.4747 15.81 3300 0.5497 0.8763 0.8838 0.8763 0.8758 0.9139
0.4456 16.05 3350 0.5757 0.8838 0.8896 0.8838 0.8835 0.9192
0.4456 16.29 3400 0.5547 0.8756 0.8830 0.8756 0.8731 0.9123
0.4456 16.53 3450 0.5431 0.8808 0.8883 0.8808 0.8801 0.9168
0.4456 16.77 3500 0.5459 0.8823 0.8883 0.8823 0.8815 0.9175
0.4248 17.01 3550 0.5111 0.8891 0.8947 0.8891 0.8878 0.9220
0.4248 17.25 3600 0.5371 0.8868 0.8922 0.8868 0.8860 0.9207
0.4248 17.49 3650 0.5757 0.8748 0.8843 0.8748 0.8745 0.9131
0.4248 17.72 3700 0.5509 0.8816 0.8880 0.8816 0.8804 0.9168
0.4248 17.96 3750 0.5166 0.8853 0.8911 0.8853 0.8845 0.9197
0.405 18.2 3800 0.5392 0.8823 0.8881 0.8823 0.8814 0.9173
0.405 18.44 3850 0.5357 0.8793 0.8857 0.8793 0.8784 0.9155
0.405 18.68 3900 0.5564 0.8748 0.8808 0.8748 0.8739 0.9120
0.405 18.92 3950 0.5377 0.8853 0.8898 0.8853 0.8842 0.9202
0.3925 19.16 4000 0.5489 0.8846 0.8902 0.8846 0.8832 0.9194
0.3925 19.4 4050 0.5953 0.8726 0.8800 0.8726 0.8713 0.9115
0.3925 19.64 4100 0.5802 0.8756 0.8812 0.8756 0.8738 0.9131
0.3925 19.88 4150 0.6130 0.8756 0.8827 0.8756 0.8743 0.9121
0.3707 20.12 4200 0.6210 0.8771 0.8828 0.8771 0.8760 0.9137
0.3707 20.36 4250 0.6460 0.8786 0.8849 0.8786 0.8774 0.9154
0.3707 20.6 4300 0.6255 0.8703 0.8780 0.8703 0.8694 0.9085
0.3707 20.84 4350 0.6773 0.8658 0.8739 0.8658 0.8653 0.9056

Framework versions

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