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
1113d9f verified
preview code
|
raw
history blame
14.6 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: 0.5653
  • Accuracy: 0.8814
  • Precision: 0.8996
  • Recall: 0.8814
  • F1: 0.8797
  • Binary: 0.9213

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 4.4122 0.0270 0.0009 0.0270 0.0018 0.1908
No log 0.38 100 3.8703 0.0485 0.0031 0.0485 0.0057 0.3057
No log 0.58 150 3.4529 0.0728 0.0261 0.0728 0.0208 0.3396
No log 0.77 200 3.3007 0.0701 0.0092 0.0701 0.0156 0.3385
No log 0.96 250 3.2150 0.0809 0.0227 0.0809 0.0284 0.3477
3.8633 1.15 300 3.1101 0.0997 0.0221 0.0997 0.0330 0.3588
3.8633 1.34 350 3.0269 0.1132 0.0298 0.1132 0.0427 0.3704
3.8633 1.53 400 2.9362 0.1509 0.0640 0.1509 0.0799 0.3943
3.8633 1.73 450 2.7990 0.2183 0.1378 0.2183 0.1356 0.4488
3.8633 1.92 500 2.6316 0.2345 0.1408 0.2345 0.1447 0.4609
3.1477 2.11 550 2.4207 0.2911 0.1703 0.2911 0.1919 0.5027
3.1477 2.3 600 2.2241 0.3315 0.2375 0.3315 0.2437 0.5329
3.1477 2.49 650 2.0490 0.4016 0.3209 0.4016 0.3144 0.5782
3.1477 2.68 700 1.9507 0.4474 0.3860 0.4474 0.3751 0.6111
3.1477 2.88 750 1.7289 0.4987 0.4422 0.4987 0.4249 0.6469
2.5128 3.07 800 1.6980 0.5229 0.4922 0.5229 0.4716 0.6650
2.5128 3.26 850 1.5772 0.5337 0.4984 0.5337 0.4794 0.6752
2.5128 3.45 900 1.4601 0.5526 0.5254 0.5526 0.4960 0.6857
2.5128 3.64 950 1.3596 0.5984 0.5985 0.5984 0.5666 0.7175
2.5128 3.84 1000 1.2260 0.6550 0.7094 0.6550 0.6282 0.7580
2.0178 4.03 1050 1.2179 0.6334 0.6586 0.6334 0.5903 0.7429
2.0178 4.22 1100 1.1750 0.6523 0.6673 0.6523 0.6209 0.7569
2.0178 4.41 1150 1.1022 0.6927 0.7149 0.6927 0.6702 0.7852
2.0178 4.6 1200 1.0216 0.7143 0.7344 0.7143 0.6957 0.8005
2.0178 4.79 1250 1.0195 0.7008 0.7571 0.7008 0.6843 0.7908
2.0178 4.99 1300 1.0202 0.6765 0.6885 0.6765 0.6532 0.7739
1.715 5.18 1350 0.9848 0.7547 0.7772 0.7547 0.7378 0.8264
1.715 5.37 1400 1.0069 0.7385 0.7797 0.7385 0.7229 0.8146
1.715 5.56 1450 0.9229 0.7547 0.7937 0.7547 0.7420 0.8267
1.715 5.75 1500 0.8889 0.7385 0.7600 0.7385 0.7170 0.8191
1.715 5.94 1550 0.8469 0.7736 0.8162 0.7736 0.7670 0.8410
1.5024 6.14 1600 0.8843 0.7844 0.8278 0.7844 0.7767 0.8485
1.5024 6.33 1650 0.8125 0.7898 0.8149 0.7898 0.7797 0.8512
1.5024 6.52 1700 0.9072 0.7520 0.7806 0.7520 0.7365 0.8240
1.5024 6.71 1750 0.7896 0.7547 0.7818 0.7547 0.7398 0.8267
1.5024 6.9 1800 0.7102 0.8113 0.8396 0.8113 0.8045 0.8663
1.3605 7.09 1850 0.7241 0.8032 0.8298 0.8032 0.7930 0.8606
1.3605 7.29 1900 0.7586 0.8005 0.8323 0.8005 0.7892 0.8588
1.3605 7.48 1950 0.7546 0.7898 0.8105 0.7898 0.7759 0.8512
1.3605 7.67 2000 0.7103 0.8032 0.8242 0.8032 0.7893 0.8606
1.3605 7.86 2050 0.7397 0.8140 0.8239 0.8140 0.7982 0.8682
1.2475 8.05 2100 0.7723 0.8059 0.8207 0.8059 0.7894 0.8636
1.2475 8.25 2150 0.7099 0.7925 0.8088 0.7925 0.7790 0.8571
1.2475 8.44 2200 0.6816 0.8167 0.8316 0.8167 0.7998 0.8712
1.2475 8.63 2250 0.6676 0.8113 0.8264 0.8113 0.8025 0.8663
1.2475 8.82 2300 0.7176 0.8113 0.8326 0.8113 0.7973 0.8663
1.1791 9.01 2350 0.6161 0.8356 0.8500 0.8356 0.8238 0.8833
1.1791 9.2 2400 0.6973 0.8032 0.8165 0.8032 0.7873 0.8606
1.1791 9.4 2450 0.6981 0.8248 0.8531 0.8248 0.8138 0.8757
1.1791 9.59 2500 0.6134 0.8356 0.8417 0.8356 0.8231 0.8833
1.1791 9.78 2550 0.5840 0.8356 0.8498 0.8356 0.8220 0.8833
1.1791 9.97 2600 0.5940 0.8329 0.8479 0.8329 0.8170 0.8814
1.1047 10.16 2650 0.5539 0.8491 0.8641 0.8491 0.8435 0.8935
1.1047 10.35 2700 0.6522 0.8410 0.8582 0.8410 0.8269 0.8871
1.1047 10.55 2750 0.5940 0.8356 0.8631 0.8356 0.8263 0.8852
1.1047 10.74 2800 0.5980 0.8464 0.8558 0.8464 0.8376 0.8927
1.1047 10.93 2850 0.6173 0.8302 0.8518 0.8302 0.8222 0.8795
1.0458 11.12 2900 0.5973 0.8356 0.8580 0.8356 0.8257 0.8833
1.0458 11.31 2950 0.5135 0.8706 0.8853 0.8706 0.8660 0.9078
1.0458 11.51 3000 0.5858 0.8410 0.8551 0.8410 0.8290 0.8871
1.0458 11.7 3050 0.6788 0.8248 0.8479 0.8248 0.8113 0.8757
1.0458 11.89 3100 0.5917 0.8437 0.8565 0.8437 0.8306 0.8889
0.9866 12.08 3150 0.6466 0.8194 0.8328 0.8194 0.8078 0.8720
0.9866 12.27 3200 0.6311 0.8194 0.8357 0.8194 0.8102 0.8720
0.9866 12.46 3250 0.6292 0.8383 0.8589 0.8383 0.8276 0.8852
0.9866 12.66 3300 0.5887 0.8437 0.8609 0.8437 0.8360 0.8908
0.9866 12.85 3350 0.6003 0.8302 0.8490 0.8302 0.8221 0.8795
0.9574 13.04 3400 0.5590 0.8625 0.8881 0.8625 0.8580 0.9022
0.9574 13.23 3450 0.6750 0.8086 0.8223 0.8086 0.7958 0.8644
0.9574 13.42 3500 0.6180 0.8302 0.8556 0.8302 0.8245 0.8795
0.9574 13.61 3550 0.5702 0.8625 0.8812 0.8625 0.8572 0.9022
0.9574 13.81 3600 0.5661 0.8625 0.8747 0.8625 0.8536 0.9022
0.9574 14.0 3650 0.6820 0.8302 0.8449 0.8302 0.8230 0.8825
0.9173 14.19 3700 0.5872 0.8544 0.8772 0.8544 0.8474 0.8984
0.9173 14.38 3750 0.5503 0.8571 0.8758 0.8571 0.8506 0.8984
0.9173 14.57 3800 0.5711 0.8652 0.8889 0.8652 0.8594 0.9040
0.9173 14.77 3850 0.5832 0.8491 0.8703 0.8491 0.8431 0.8927
0.9173 14.96 3900 0.5457 0.8706 0.8929 0.8706 0.8658 0.9089
0.8859 15.15 3950 0.6410 0.8491 0.8667 0.8491 0.8406 0.8927
0.8859 15.34 4000 0.5822 0.8410 0.8661 0.8410 0.8340 0.8871
0.8859 15.53 4050 0.6173 0.8464 0.8720 0.8464 0.8406 0.8919
0.8859 15.72 4100 0.6509 0.8356 0.8535 0.8356 0.8267 0.8833
0.8859 15.92 4150 0.7177 0.8275 0.8419 0.8275 0.8156 0.8776
0.8447 16.11 4200 0.5898 0.8437 0.8531 0.8437 0.8347 0.8889
0.8447 16.3 4250 0.6429 0.8383 0.8513 0.8383 0.8296 0.8852
0.8447 16.49 4300 0.5914 0.8625 0.8707 0.8625 0.8553 0.9022
0.8447 16.68 4350 0.5698 0.8518 0.8714 0.8518 0.8460 0.8946
0.8447 16.87 4400 0.5938 0.8491 0.8695 0.8491 0.8439 0.8946
0.8181 17.07 4450 0.6076 0.8356 0.8441 0.8356 0.8263 0.8852
0.8181 17.26 4500 0.5691 0.8383 0.8518 0.8383 0.8298 0.8852
0.8181 17.45 4550 0.5490 0.8625 0.8737 0.8625 0.8551 0.9040
0.8181 17.64 4600 0.5963 0.8598 0.8791 0.8598 0.8538 0.9003
0.8181 17.83 4650 0.6371 0.8464 0.8699 0.8464 0.8406 0.8908
0.8015 18.02 4700 0.6348 0.8491 0.8675 0.8491 0.8449 0.8927
0.8015 18.22 4750 0.6207 0.8571 0.8711 0.8571 0.8487 0.8984
0.8015 18.41 4800 0.6759 0.8518 0.8709 0.8518 0.8479 0.8946
0.8015 18.6 4850 0.7267 0.8248 0.8346 0.8248 0.8136 0.8757
0.8015 18.79 4900 0.6420 0.8410 0.8629 0.8410 0.8361 0.8871
0.8015 18.98 4950 0.6260 0.8464 0.8581 0.8464 0.8375 0.8908
0.7757 19.18 5000 0.6714 0.8410 0.8666 0.8410 0.8361 0.8889
0.7757 19.37 5050 0.6414 0.8383 0.8485 0.8383 0.8285 0.8852
0.7757 19.56 5100 0.6348 0.8356 0.8547 0.8356 0.8261 0.8833
0.7757 19.75 5150 0.6811 0.8464 0.8625 0.8464 0.8377 0.8908
0.7757 19.94 5200 0.6294 0.8383 0.8511 0.8383 0.8286 0.8852
0.7456 20.13 5250 0.6511 0.8679 0.8785 0.8679 0.8589 0.9078
0.7456 20.33 5300 0.6374 0.8437 0.8543 0.8437 0.8344 0.8889
0.7456 20.52 5350 0.6019 0.8544 0.8648 0.8544 0.8457 0.8965
0.7456 20.71 5400 0.6060 0.8571 0.8632 0.8571 0.8469 0.8984
0.7456 20.9 5450 0.6730 0.8518 0.8626 0.8518 0.8453 0.8946
0.7406 21.09 5500 0.6091 0.8544 0.8633 0.8544 0.8450 0.8965
0.7406 21.28 5550 0.6378 0.8598 0.8691 0.8598 0.8511 0.9003
0.7406 21.48 5600 0.5868 0.8464 0.8543 0.8464 0.8388 0.8908
0.7406 21.67 5650 0.5930 0.8706 0.8864 0.8706 0.8658 0.9078
0.7406 21.86 5700 0.6086 0.8544 0.8711 0.8544 0.8497 0.8965
0.7057 22.05 5750 0.6130 0.8518 0.8751 0.8518 0.8471 0.8946
0.7057 22.24 5800 0.6477 0.8464 0.8728 0.8464 0.8393 0.8908
0.7057 22.44 5850 0.6165 0.8518 0.8595 0.8518 0.8434 0.8946
0.7057 22.63 5900 0.6288 0.8571 0.8693 0.8571 0.8491 0.8984
0.7057 22.82 5950 0.6246 0.8544 0.8749 0.8544 0.8490 0.8965
0.695 23.01 6000 0.5991 0.8679 0.8874 0.8679 0.8645 0.9059
0.695 23.2 6050 0.6234 0.8598 0.8816 0.8598 0.8556 0.9003
0.695 23.39 6100 0.5764 0.8679 0.8885 0.8679 0.8641 0.9059
0.695 23.59 6150 0.6290 0.8518 0.8641 0.8518 0.8453 0.8946
0.695 23.78 6200 0.6267 0.8518 0.8634 0.8518 0.8433 0.8946
0.695 23.97 6250 0.6294 0.8491 0.8582 0.8491 0.8404 0.8927
0.6782 24.16 6300 0.6001 0.8491 0.8618 0.8491 0.8421 0.8927
0.6782 24.35 6350 0.6042 0.8598 0.8687 0.8598 0.8530 0.9003

Framework versions

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