Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-1-old

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-1-old / README.md
fydhfzh's picture
fydhfzh
End of training
58d5cf1 verified
preview code
|
raw
history blame
13.6 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-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: 0.5721
  • Accuracy: 0.8706
  • Precision: 0.8813
  • Recall: 0.8706
  • F1: 0.8644
  • Binary: 0.9094

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 0.19 50 4.4259 0.0135 0.0002 0.0135 0.0004 0.1278
No log 0.38 100 3.8225 0.0485 0.0033 0.0485 0.0060 0.3151
No log 0.58 150 3.3928 0.0701 0.0104 0.0701 0.0169 0.3450
No log 0.77 200 3.2088 0.0836 0.0166 0.0836 0.0258 0.3553
No log 0.96 250 3.0752 0.1078 0.0279 0.1078 0.0388 0.3714
3.8204 1.15 300 2.9617 0.1186 0.0338 0.1186 0.0443 0.3776
3.8204 1.34 350 2.7928 0.1617 0.0704 0.1617 0.0848 0.4100
3.8204 1.53 400 2.5904 0.2237 0.1640 0.2237 0.1427 0.4542
3.8204 1.73 450 2.3895 0.2615 0.1634 0.2615 0.1771 0.4814
3.8204 1.92 500 2.2567 0.2992 0.2170 0.2992 0.2160 0.5097
2.9288 2.11 550 2.0903 0.3801 0.2993 0.3801 0.3008 0.5668
2.9288 2.3 600 1.9624 0.4151 0.3638 0.4151 0.3389 0.5900
2.9288 2.49 650 1.7353 0.5148 0.4641 0.5148 0.4447 0.6625
2.9288 2.68 700 1.6687 0.5013 0.4658 0.5013 0.4433 0.6526
2.9288 2.88 750 1.5726 0.5391 0.5299 0.5391 0.4914 0.6817
2.2908 3.07 800 1.4471 0.5714 0.5785 0.5714 0.5269 0.7003
2.2908 3.26 850 1.3350 0.6334 0.6432 0.6334 0.6040 0.7445
2.2908 3.45 900 1.1787 0.6685 0.6811 0.6685 0.6349 0.7690
2.2908 3.64 950 1.1315 0.6846 0.6962 0.6846 0.6524 0.7803
2.2908 3.84 1000 1.0283 0.7493 0.7751 0.7493 0.7298 0.8248
1.8601 4.03 1050 1.0553 0.7089 0.7311 0.7089 0.6824 0.7984
1.8601 4.22 1100 0.9262 0.7385 0.7627 0.7385 0.7190 0.8170
1.8601 4.41 1150 0.9209 0.7385 0.7586 0.7385 0.7221 0.8170
1.8601 4.6 1200 0.9163 0.7520 0.8027 0.7520 0.7416 0.8280
1.8601 4.79 1250 0.7954 0.8140 0.8394 0.8140 0.8028 0.8682
1.8601 4.99 1300 0.8244 0.7709 0.7999 0.7709 0.7651 0.8388
1.624 5.18 1350 0.8025 0.7655 0.8100 0.7655 0.7599 0.8364
1.624 5.37 1400 0.7521 0.7925 0.8238 0.7925 0.7827 0.8558
1.624 5.56 1450 0.7821 0.7898 0.8231 0.7898 0.7824 0.8528
1.624 5.75 1500 0.6841 0.7951 0.8237 0.7951 0.7887 0.8577
1.624 5.94 1550 0.6757 0.7978 0.8276 0.7978 0.7935 0.8596
1.4343 6.14 1600 0.6709 0.8140 0.8415 0.8140 0.8094 0.8709
1.4343 6.33 1650 0.6361 0.8113 0.8364 0.8113 0.8045 0.8690
1.4343 6.52 1700 0.6413 0.8275 0.8479 0.8275 0.8231 0.8814
1.4343 6.71 1750 0.6074 0.8302 0.8484 0.8302 0.8244 0.8803
1.4343 6.9 1800 0.6286 0.8005 0.8321 0.8005 0.7964 0.8606
1.3175 7.09 1850 0.5431 0.8356 0.8558 0.8356 0.8312 0.8860
1.3175 7.29 1900 0.5612 0.8491 0.8828 0.8491 0.8499 0.8927
1.3175 7.48 1950 0.5324 0.8491 0.8795 0.8491 0.8492 0.8954
1.3175 7.67 2000 0.5793 0.8383 0.8589 0.8383 0.8345 0.8871
1.3175 7.86 2050 0.5722 0.8248 0.8575 0.8248 0.8258 0.8757
1.2154 8.05 2100 0.6362 0.8167 0.8511 0.8167 0.8149 0.8701
1.2154 8.25 2150 0.5846 0.8302 0.8588 0.8302 0.8293 0.8814
1.2154 8.44 2200 0.6121 0.8113 0.8477 0.8113 0.8065 0.8704
1.2154 8.63 2250 0.5895 0.8356 0.8671 0.8356 0.8343 0.8871
1.2154 8.82 2300 0.5404 0.8437 0.8756 0.8437 0.8407 0.8927
1.1306 9.01 2350 0.5433 0.8410 0.8657 0.8410 0.8394 0.8900
1.1306 9.2 2400 0.5535 0.8383 0.8645 0.8383 0.8380 0.8881
1.1306 9.4 2450 0.5201 0.8518 0.8850 0.8518 0.8518 0.8957
1.1306 9.59 2500 0.5464 0.8383 0.8680 0.8383 0.8373 0.8881
1.1306 9.78 2550 0.5960 0.8329 0.8586 0.8329 0.8304 0.8841
1.1306 9.97 2600 0.5304 0.8518 0.8790 0.8518 0.8506 0.8957
1.0743 10.16 2650 0.4804 0.8706 0.8937 0.8706 0.8703 0.9086
1.0743 10.35 2700 0.5004 0.8652 0.8908 0.8652 0.8640 0.9059
1.0743 10.55 2750 0.4730 0.8652 0.8921 0.8652 0.8643 0.9070
1.0743 10.74 2800 0.4958 0.8383 0.8663 0.8383 0.8372 0.8879
1.0743 10.93 2850 0.4672 0.8544 0.8814 0.8544 0.8538 0.8973
1.0274 11.12 2900 0.5339 0.8571 0.8807 0.8571 0.8565 0.9003
1.0274 11.31 2950 0.5013 0.8491 0.8698 0.8491 0.8462 0.8954
1.0274 11.51 3000 0.4882 0.8679 0.8904 0.8679 0.8677 0.9078
1.0274 11.7 3050 0.5059 0.8518 0.8837 0.8518 0.8519 0.8965
1.0274 11.89 3100 0.4636 0.8679 0.8864 0.8679 0.8666 0.9075
0.9585 12.08 3150 0.4667 0.8787 0.8955 0.8787 0.8776 0.9143
0.9585 12.27 3200 0.5159 0.8544 0.8734 0.8544 0.8534 0.8976
0.9585 12.46 3250 0.5177 0.8518 0.8748 0.8518 0.8528 0.8987
0.9585 12.66 3300 0.4435 0.8841 0.9040 0.8841 0.8835 0.9189
0.9585 12.85 3350 0.5116 0.8544 0.8851 0.8544 0.8558 0.8992
0.9352 13.04 3400 0.4538 0.8706 0.8888 0.8706 0.8697 0.9105
0.9352 13.23 3450 0.4973 0.8706 0.8944 0.8706 0.8684 0.9086
0.9352 13.42 3500 0.4465 0.8760 0.8937 0.8760 0.8741 0.9135
0.9352 13.61 3550 0.4691 0.8814 0.9042 0.8814 0.8806 0.9154
0.9352 13.81 3600 0.5010 0.8652 0.8916 0.8652 0.8641 0.9051
0.9352 14.0 3650 0.5133 0.8410 0.8728 0.8410 0.8396 0.8879
0.8941 14.19 3700 0.4476 0.8706 0.8961 0.8706 0.8729 0.9086
0.8941 14.38 3750 0.4321 0.8679 0.8915 0.8679 0.8681 0.9067
0.8941 14.57 3800 0.4033 0.8841 0.8991 0.8841 0.8835 0.9181
0.8941 14.77 3850 0.4599 0.8841 0.9052 0.8841 0.8827 0.9181
0.8941 14.96 3900 0.4673 0.8625 0.8883 0.8625 0.8631 0.9040
0.8574 15.15 3950 0.4906 0.8760 0.8993 0.8760 0.8749 0.9135
0.8574 15.34 4000 0.5055 0.8544 0.8836 0.8544 0.8518 0.8984
0.8574 15.53 4050 0.4119 0.8841 0.8985 0.8841 0.8831 0.9191
0.8574 15.72 4100 0.4684 0.8760 0.8989 0.8760 0.8752 0.9135
0.8574 15.92 4150 0.4453 0.8787 0.8999 0.8787 0.8776 0.9151
0.8344 16.11 4200 0.4928 0.8787 0.9000 0.8787 0.8783 0.9143
0.8344 16.3 4250 0.4535 0.8868 0.9067 0.8868 0.8863 0.9191
0.8344 16.49 4300 0.4259 0.8787 0.8986 0.8787 0.8781 0.9154
0.8344 16.68 4350 0.4289 0.8787 0.8970 0.8787 0.8784 0.9154
0.8344 16.87 4400 0.4828 0.8814 0.9013 0.8814 0.8813 0.9154
0.8066 17.07 4450 0.4866 0.8787 0.8969 0.8787 0.8792 0.9135
0.8066 17.26 4500 0.4388 0.8760 0.8934 0.8760 0.8768 0.9116
0.8066 17.45 4550 0.5018 0.8787 0.8993 0.8787 0.8775 0.9143
0.8066 17.64 4600 0.4838 0.8814 0.8981 0.8814 0.8808 0.9154
0.8066 17.83 4650 0.5394 0.8679 0.8893 0.8679 0.8662 0.9059
0.7757 18.02 4700 0.4628 0.8814 0.8964 0.8814 0.8800 0.9162
0.7757 18.22 4750 0.5456 0.8733 0.8907 0.8733 0.8719 0.9097
0.7757 18.41 4800 0.4858 0.8814 0.8970 0.8814 0.8804 0.9154
0.7757 18.6 4850 0.5836 0.8571 0.8776 0.8571 0.8568 0.8984
0.7757 18.79 4900 0.5008 0.8787 0.8985 0.8787 0.8781 0.9143
0.7757 18.98 4950 0.5259 0.8760 0.8950 0.8760 0.8749 0.9116
0.7595 19.18 5000 0.5906 0.8652 0.8869 0.8652 0.8657 0.9040
0.7595 19.37 5050 0.4905 0.8841 0.8993 0.8841 0.8839 0.9173
0.7595 19.56 5100 0.5958 0.8598 0.8804 0.8598 0.8596 0.9003
0.7595 19.75 5150 0.5466 0.8679 0.8924 0.8679 0.8666 0.9059
0.7595 19.94 5200 0.4639 0.8841 0.9008 0.8841 0.8834 0.9173
0.7257 20.13 5250 0.5094 0.8787 0.9015 0.8787 0.8795 0.9135
0.7257 20.33 5300 0.5310 0.8733 0.8973 0.8733 0.8737 0.9097
0.7257 20.52 5350 0.5118 0.8733 0.8925 0.8733 0.8734 0.9097
0.7257 20.71 5400 0.5166 0.8814 0.9017 0.8814 0.8814 0.9154
0.7257 20.9 5450 0.4850 0.8814 0.8984 0.8814 0.8807 0.9164
0.7185 21.09 5500 0.5161 0.8841 0.9018 0.8841 0.8842 0.9183
0.7185 21.28 5550 0.5197 0.8706 0.8904 0.8706 0.8694 0.9086
0.7185 21.48 5600 0.5297 0.8733 0.8921 0.8733 0.8728 0.9097
0.7185 21.67 5650 0.5317 0.8706 0.8913 0.8706 0.8694 0.9078
0.7185 21.86 5700 0.5120 0.8625 0.8809 0.8625 0.8610 0.9022
0.6968 22.05 5750 0.5144 0.8760 0.8960 0.8760 0.8753 0.9116
0.6968 22.24 5800 0.5688 0.8733 0.8911 0.8733 0.8721 0.9097
0.6968 22.44 5850 0.5430 0.8733 0.8916 0.8733 0.8724 0.9097

Framework versions

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