Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-3

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-3 / README.md
fydhfzh's picture
fydhfzh
End of training
099e8d1 verified
preview code
|
raw
history blame
11.9 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-3
    results: []

hubert-classifier-aug-fold-3

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: 0.6680
  • Accuracy: 0.8787
  • Precision: 0.8938
  • Recall: 0.8787
  • F1: 0.8770
  • Binary: 0.9158

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 0.13 50 4.4184 0.0243 0.0045 0.0243 0.0062 0.1841
No log 0.27 100 4.2655 0.0418 0.0031 0.0418 0.0053 0.2968
No log 0.4 150 3.9561 0.0472 0.0055 0.0472 0.0083 0.3246
No log 0.54 200 3.6420 0.0823 0.0299 0.0823 0.0317 0.3511
No log 0.67 250 3.3980 0.1201 0.0645 0.1201 0.0588 0.3811
No log 0.81 300 3.1749 0.1957 0.1344 0.1957 0.1165 0.4321
No log 0.94 350 2.9362 0.2618 0.1776 0.2618 0.1713 0.4819
3.8313 1.08 400 2.6613 0.3347 0.2471 0.3347 0.2417 0.5328
3.8313 1.21 450 2.3242 0.4130 0.3415 0.4130 0.3296 0.5868
3.8313 1.35 500 2.0383 0.4858 0.4259 0.4858 0.4186 0.6387
3.8313 1.48 550 1.7209 0.5776 0.5077 0.5776 0.5093 0.7053
3.8313 1.62 600 1.4932 0.6019 0.5524 0.6019 0.5396 0.7238
3.8313 1.75 650 1.3401 0.6667 0.6714 0.6667 0.6328 0.7664
3.8313 1.89 700 1.1896 0.7139 0.7150 0.7139 0.6812 0.7992
2.0747 2.02 750 1.0789 0.7099 0.7139 0.7099 0.6800 0.7953
2.0747 2.16 800 1.0023 0.7463 0.7432 0.7463 0.7197 0.8238
2.0747 2.29 850 0.9637 0.7503 0.7619 0.7503 0.7348 0.8262
2.0747 2.43 900 0.9403 0.7544 0.7749 0.7544 0.7398 0.8279
2.0747 2.56 950 0.8283 0.7773 0.7827 0.7773 0.7665 0.8439
2.0747 2.7 1000 0.8475 0.7908 0.8127 0.7908 0.7808 0.8521
2.0747 2.83 1050 0.8050 0.7611 0.7921 0.7611 0.7570 0.8328
2.0747 2.96 1100 0.7450 0.8111 0.8226 0.8111 0.8019 0.8686
1.1049 3.1 1150 0.7468 0.8165 0.8359 0.8165 0.8145 0.8709
1.1049 3.23 1200 0.7577 0.8057 0.8333 0.8057 0.7980 0.8648
1.1049 3.37 1250 0.7135 0.8273 0.8392 0.8273 0.8234 0.8799
1.1049 3.5 1300 0.7512 0.8124 0.8269 0.8124 0.8084 0.8690
1.1049 3.64 1350 0.7234 0.8192 0.8377 0.8192 0.8128 0.8740
1.1049 3.77 1400 0.6902 0.8219 0.8394 0.8219 0.8181 0.8752
1.1049 3.91 1450 0.7227 0.8111 0.8208 0.8111 0.8045 0.8676
0.755 4.04 1500 0.6752 0.8273 0.8458 0.8273 0.8227 0.8788
0.755 4.18 1550 0.6767 0.8300 0.8440 0.8300 0.8248 0.8807
0.755 4.31 1600 0.7044 0.8192 0.8352 0.8192 0.8132 0.8737
0.755 4.45 1650 0.7419 0.8246 0.8502 0.8246 0.8215 0.8776
0.755 4.58 1700 0.7255 0.8192 0.8429 0.8192 0.8153 0.8742
0.755 4.72 1750 0.7030 0.8354 0.8586 0.8354 0.8332 0.8860
0.755 4.85 1800 0.6936 0.8421 0.8640 0.8421 0.8394 0.8901
0.755 4.99 1850 0.6561 0.8394 0.8586 0.8394 0.8351 0.8883
0.5857 5.12 1900 0.7286 0.8381 0.8604 0.8381 0.8344 0.8874
0.5857 5.26 1950 0.6338 0.8462 0.8645 0.8462 0.8444 0.8939
0.5857 5.39 2000 0.6636 0.8408 0.8599 0.8408 0.8392 0.8888
0.5857 5.53 2050 0.7965 0.8246 0.8483 0.8246 0.8223 0.8789
0.5857 5.66 2100 0.6798 0.8475 0.8655 0.8475 0.8439 0.8949
0.5857 5.8 2150 0.6231 0.8529 0.8687 0.8529 0.8508 0.8973
0.5857 5.93 2200 0.6443 0.8435 0.8587 0.8435 0.8397 0.8916
0.4922 6.06 2250 0.6697 0.8394 0.8607 0.8394 0.8378 0.8883
0.4922 6.2 2300 0.7305 0.8259 0.8553 0.8259 0.8258 0.8779
0.4922 6.33 2350 0.6740 0.8502 0.8682 0.8502 0.8495 0.8953
0.4922 6.47 2400 0.7172 0.8421 0.8641 0.8421 0.8403 0.8892
0.4922 6.6 2450 0.7219 0.8556 0.8727 0.8556 0.8495 0.8989
0.4922 6.74 2500 0.6512 0.8502 0.8657 0.8502 0.8479 0.8954
0.4922 6.87 2550 0.6893 0.8408 0.8594 0.8408 0.8372 0.8901
0.4266 7.01 2600 0.6625 0.8475 0.8630 0.8475 0.8458 0.8943
0.4266 7.14 2650 0.8103 0.8246 0.8526 0.8246 0.8239 0.8771
0.4266 7.28 2700 0.7695 0.8556 0.8755 0.8556 0.8534 0.8997
0.4266 7.41 2750 0.7239 0.8340 0.8581 0.8340 0.8332 0.8846
0.4266 7.55 2800 0.7330 0.8340 0.8553 0.8340 0.8315 0.8850
0.4266 7.68 2850 0.7010 0.8516 0.8746 0.8516 0.8504 0.8973
0.4266 7.82 2900 0.7827 0.8421 0.8627 0.8421 0.8400 0.8912
0.4266 7.95 2950 0.6885 0.8502 0.8659 0.8502 0.8495 0.8964
0.3842 8.09 3000 0.7856 0.8475 0.8690 0.8475 0.8454 0.8939
0.3842 8.22 3050 0.8063 0.8354 0.8597 0.8354 0.8331 0.8853
0.3842 8.36 3100 0.6893 0.8610 0.8726 0.8610 0.8581 0.9028
0.3842 8.49 3150 0.7546 0.8462 0.8643 0.8462 0.8444 0.8934
0.3842 8.63 3200 0.7635 0.8489 0.8712 0.8489 0.8474 0.8958
0.3842 8.76 3250 0.7346 0.8462 0.8626 0.8462 0.8441 0.8930
0.3842 8.89 3300 0.8108 0.8529 0.8629 0.8529 0.8474 0.8978
0.342 9.03 3350 0.6884 0.8637 0.8767 0.8637 0.8621 0.9062
0.342 9.16 3400 0.7026 0.8704 0.8856 0.8704 0.8698 0.9100
0.342 9.3 3450 0.7660 0.8489 0.8674 0.8489 0.8464 0.8943
0.342 9.43 3500 0.7238 0.8570 0.8738 0.8570 0.8558 0.9016
0.342 9.57 3550 0.7542 0.8677 0.8820 0.8677 0.8666 0.9090
0.342 9.7 3600 0.7187 0.8489 0.8668 0.8489 0.8483 0.8949
0.342 9.84 3650 0.6752 0.8664 0.8794 0.8664 0.8645 0.9067
0.342 9.97 3700 0.7404 0.8637 0.8809 0.8637 0.8617 0.9053
0.3175 10.11 3750 0.7444 0.8623 0.8831 0.8623 0.8619 0.9035
0.3175 10.24 3800 0.7315 0.8543 0.8744 0.8543 0.8528 0.8987
0.3175 10.38 3850 0.7424 0.8529 0.8730 0.8529 0.8519 0.8969
0.3175 10.51 3900 0.6655 0.8664 0.8820 0.8664 0.8649 0.9076
0.3175 10.65 3950 0.7943 0.8570 0.8775 0.8570 0.8553 0.9005
0.3175 10.78 4000 0.7559 0.8583 0.8790 0.8583 0.8579 0.9019
0.3175 10.92 4050 0.7496 0.8489 0.8650 0.8489 0.8475 0.8964
0.2912 11.05 4100 0.7507 0.8570 0.8731 0.8570 0.8552 0.9005
0.2912 11.19 4150 0.7952 0.8596 0.8776 0.8596 0.8588 0.9034
0.2912 11.32 4200 0.7547 0.8516 0.8668 0.8516 0.8499 0.8977
0.2912 11.46 4250 0.8149 0.8475 0.8677 0.8475 0.8442 0.8954
0.2912 11.59 4300 0.7429 0.8596 0.8741 0.8596 0.8577 0.9023
0.2912 11.73 4350 0.7403 0.8650 0.8810 0.8650 0.8635 0.9062
0.2912 11.86 4400 0.7918 0.8570 0.8705 0.8570 0.8538 0.9009
0.2912 11.99 4450 0.7712 0.8610 0.8773 0.8610 0.8586 0.9043
0.2795 12.13 4500 0.7388 0.8677 0.8814 0.8677 0.8650 0.9096
0.2795 12.26 4550 0.7508 0.8677 0.8814 0.8677 0.8670 0.9090
0.2795 12.4 4600 0.8635 0.8543 0.8730 0.8543 0.8513 0.8987
0.2795 12.53 4650 0.7977 0.8677 0.8831 0.8677 0.8656 0.9090
0.2795 12.67 4700 0.7686 0.8556 0.8750 0.8556 0.8562 0.8996
0.2795 12.8 4750 0.7998 0.8570 0.8719 0.8570 0.8557 0.9015
0.2795 12.94 4800 0.8172 0.8637 0.8762 0.8637 0.8617 0.9057
0.2638 13.07 4850 0.8317 0.8502 0.8670 0.8502 0.8481 0.8968
0.2638 13.21 4900 0.8888 0.8529 0.8664 0.8529 0.8516 0.8977
0.2638 13.34 4950 0.8767 0.8583 0.8763 0.8583 0.8570 0.9011
0.2638 13.48 5000 0.8237 0.8610 0.8762 0.8610 0.8607 0.9028

Framework versions

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