Hugging Face's logo

fydhfzh
/
wav2vec2-classifier

Audio Classification
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Model card Files Metrics Community
wav2vec2-classifier / README.md
fydhfzh's picture
fydhfzh
End of training
43be8f5 verified
preview code
|
raw
history blame
7.47 kB
metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: wav2vec2-classifier
    results: []

wav2vec2-classifier

This model is a fine-tuned version of facebook/wav2vec2-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8709
  • Accuracy: 0.8083
  • Precision: 0.8377
  • Recall: 0.8083
  • F1: 0.8018
  • Binary: 0.8650

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: 3e-05
  • 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: 10
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1 Binary
No log 0.17 50 4.2449 0.0510 0.0085 0.0510 0.0115 0.3218
No log 0.35 100 3.9731 0.0461 0.0045 0.0461 0.0074 0.3265
No log 0.52 150 3.7773 0.0437 0.0097 0.0437 0.0124 0.3301
No log 0.69 200 3.6063 0.1019 0.0598 0.1019 0.0574 0.3701
No log 0.86 250 3.4003 0.1990 0.1624 0.1990 0.1291 0.4379
3.8908 1.04 300 3.2347 0.25 0.2359 0.25 0.1908 0.4728
3.8908 1.21 350 3.0941 0.3131 0.2684 0.3131 0.2507 0.5177
3.8908 1.38 400 2.9380 0.3835 0.3642 0.3835 0.3147 0.5670
3.8908 1.55 450 2.8310 0.3956 0.3552 0.3956 0.3255 0.5750
3.8908 1.73 500 2.6928 0.4709 0.4197 0.4709 0.4024 0.6282
3.8908 1.9 550 2.6116 0.5194 0.4641 0.5194 0.4569 0.6629
3.027 2.07 600 2.4653 0.5704 0.5395 0.5704 0.5124 0.6988
3.027 2.24 650 2.3361 0.6044 0.5338 0.6044 0.5442 0.7223
3.027 2.42 700 2.2435 0.5947 0.5439 0.5947 0.5324 0.7141
3.027 2.59 750 2.1533 0.6141 0.6006 0.6141 0.5678 0.7291
3.027 2.76 800 2.0622 0.6189 0.6313 0.6189 0.5815 0.7325
3.027 2.93 850 1.9734 0.6578 0.6757 0.6578 0.6267 0.7597
2.4367 3.11 900 1.9059 0.6699 0.6688 0.6699 0.6289 0.7675
2.4367 3.28 950 1.8291 0.6990 0.7240 0.6990 0.6767 0.7886
2.4367 3.45 1000 1.7507 0.7257 0.7486 0.7257 0.6920 0.8066
2.4367 3.62 1050 1.6782 0.7330 0.7475 0.7330 0.7153 0.8124
2.4367 3.8 1100 1.6353 0.7184 0.7445 0.7184 0.6941 0.8015
2.4367 3.97 1150 1.5548 0.7694 0.7840 0.7694 0.7505 0.8379
2.0129 4.14 1200 1.5002 0.7330 0.7432 0.7330 0.7112 0.8133
2.0129 4.31 1250 1.4655 0.7524 0.7706 0.7524 0.7339 0.8260
2.0129 4.49 1300 1.4149 0.7718 0.7931 0.7718 0.7529 0.8396
2.0129 4.66 1350 1.3751 0.7743 0.7927 0.7743 0.7544 0.8422
2.0129 4.83 1400 1.3503 0.7767 0.8061 0.7767 0.7693 0.8422
1.7091 5.0 1450 1.3102 0.7816 0.8084 0.7816 0.7712 0.8473
1.7091 5.18 1500 1.3020 0.7621 0.8003 0.7621 0.7538 0.8328
1.7091 5.35 1550 1.2288 0.7816 0.8013 0.7816 0.7707 0.8473
1.7091 5.52 1600 1.2042 0.7961 0.8235 0.7961 0.7874 0.8575
1.7091 5.69 1650 1.1769 0.8010 0.8371 0.8010 0.7946 0.8592
1.7091 5.87 1700 1.1471 0.8107 0.8524 0.8107 0.8051 0.8667
1.4785 6.04 1750 1.1107 0.8228 0.8478 0.8228 0.8138 0.8745
1.4785 6.21 1800 1.0799 0.8131 0.8378 0.8131 0.8027 0.8684
1.4785 6.38 1850 1.0704 0.8010 0.8229 0.8010 0.7915 0.8600
1.4785 6.56 1900 1.0424 0.8058 0.8281 0.8058 0.7998 0.8633
1.4785 6.73 1950 1.0279 0.7985 0.8194 0.7985 0.7905 0.8592
1.4785 6.9 2000 1.0122 0.8180 0.8408 0.8180 0.8069 0.8711
1.3046 7.08 2050 0.9924 0.8180 0.8366 0.8180 0.8087 0.8711
1.3046 7.25 2100 0.9909 0.8058 0.8320 0.8058 0.7980 0.8633
1.3046 7.42 2150 0.9724 0.8058 0.8355 0.8058 0.8001 0.8643
1.3046 7.59 2200 0.9406 0.8083 0.8345 0.8083 0.8032 0.8660
1.3046 7.77 2250 0.9495 0.8010 0.8290 0.8010 0.7949 0.8592
1.3046 7.94 2300 0.9283 0.8107 0.8343 0.8107 0.8035 0.8667
1.1807 8.11 2350 0.9242 0.8155 0.8392 0.8155 0.8096 0.8711
1.1807 8.28 2400 0.9066 0.8204 0.8427 0.8204 0.8131 0.8735
1.1807 8.46 2450 0.9049 0.8252 0.8451 0.8252 0.8149 0.8769
1.1807 8.63 2500 0.9019 0.8204 0.8484 0.8204 0.8117 0.8735
1.1807 8.8 2550 0.8914 0.8131 0.8356 0.8131 0.8051 0.8684
1.1807 8.97 2600 0.8953 0.8083 0.8403 0.8083 0.8014 0.8643
1.1094 9.15 2650 0.8824 0.8107 0.8450 0.8107 0.8067 0.8667
1.1094 9.32 2700 0.8800 0.8058 0.8317 0.8058 0.7964 0.8633
1.1094 9.49 2750 0.8729 0.8180 0.8450 0.8180 0.8113 0.8718
1.1094 9.66 2800 0.8704 0.8107 0.8383 0.8107 0.8037 0.8667
1.1094 9.84 2850 0.8709 0.8083 0.8377 0.8083 0.8018 0.8650

Framework versions

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