Hugging Face's logo

fydhfzh
/
hubert-classifier-aug

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug / README.md
fydhfzh's picture
fydhfzh
End of training
878d563 verified
preview code
|
raw
history blame
7.47 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
    results: []

hubert-classifier

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: 1.1058
  • Accuracy: 0.7748
  • Precision: 0.8018
  • Recall: 0.7748
  • F1: 0.7651
  • Binary: 0.8455

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 4.2665 0.0412 0.0107 0.0412 0.0127 0.2923
No log 0.35 100 3.9427 0.0339 0.0016 0.0339 0.0030 0.3172
No log 0.52 150 3.7412 0.0363 0.0025 0.0363 0.0041 0.3206
No log 0.69 200 3.6193 0.0654 0.0238 0.0654 0.0259 0.3373
No log 0.86 250 3.4784 0.1041 0.0460 0.1041 0.0459 0.3663
No log 1.04 300 3.3705 0.1211 0.0602 0.1211 0.0466 0.3789
No log 1.21 350 3.2597 0.1768 0.0811 0.1768 0.0894 0.4218
No log 1.38 400 3.1606 0.2082 0.1867 0.2082 0.1416 0.4424
No log 1.55 450 3.0720 0.1913 0.1490 0.1913 0.1296 0.4312
3.6525 1.73 500 2.9557 0.2446 0.1432 0.2446 0.1609 0.4671
3.6525 1.9 550 2.8287 0.2857 0.2265 0.2857 0.2059 0.4973
3.6525 2.07 600 2.7005 0.3075 0.2103 0.3075 0.2154 0.5136
3.6525 2.24 650 2.6183 0.3414 0.2398 0.3414 0.2486 0.5341
3.6525 2.42 700 2.5133 0.3632 0.2942 0.3632 0.2732 0.5516
3.6525 2.59 750 2.4277 0.3753 0.3322 0.3753 0.2948 0.5615
3.6525 2.76 800 2.3329 0.4092 0.3538 0.4092 0.3338 0.5845
3.6525 2.93 850 2.2465 0.4407 0.4125 0.4407 0.3745 0.6073
3.6525 3.11 900 2.1792 0.4600 0.4329 0.4600 0.3995 0.6203
3.6525 3.28 950 2.1004 0.5109 0.4995 0.5109 0.4540 0.6550
2.6844 3.45 1000 2.0314 0.5109 0.4799 0.5109 0.4520 0.6557
2.6844 3.62 1050 1.9561 0.5400 0.5309 0.5400 0.4859 0.6743
2.6844 3.8 1100 1.9362 0.5472 0.5441 0.5472 0.5066 0.6804
2.6844 3.97 1150 1.8666 0.5642 0.5647 0.5642 0.5232 0.6930
2.6844 4.14 1200 1.8204 0.5811 0.5716 0.5811 0.5416 0.7048
2.6844 4.31 1250 1.7494 0.5908 0.6153 0.5908 0.5618 0.7109
2.6844 4.49 1300 1.6973 0.6126 0.6062 0.6126 0.5804 0.7291
2.6844 4.66 1350 1.6615 0.6053 0.5864 0.6053 0.5707 0.7211
2.6844 4.83 1400 1.6120 0.6295 0.6304 0.6295 0.6000 0.7385
2.6844 5.0 1450 1.5620 0.6610 0.6605 0.6610 0.6333 0.7615
2.1096 5.18 1500 1.5330 0.6538 0.6424 0.6538 0.6223 0.7581
2.1096 5.35 1550 1.5112 0.6707 0.6830 0.6707 0.6484 0.7707
2.1096 5.52 1600 1.4732 0.6659 0.6793 0.6659 0.6430 0.7685
2.1096 5.69 1650 1.4420 0.6755 0.6969 0.6755 0.6538 0.7734
2.1096 5.87 1700 1.4011 0.7094 0.7461 0.7094 0.6929 0.7988
2.1096 6.04 1750 1.3924 0.6780 0.6835 0.6780 0.6557 0.7760
2.1096 6.21 1800 1.3604 0.7022 0.7116 0.7022 0.6838 0.7937
2.1096 6.38 1850 1.3271 0.7070 0.7079 0.7070 0.6882 0.7954
2.1096 6.56 1900 1.3104 0.7264 0.7338 0.7264 0.7110 0.8099
2.1096 6.73 1950 1.2804 0.7312 0.7591 0.7312 0.7159 0.8131
1.7648 6.9 2000 1.2722 0.7312 0.7739 0.7312 0.7185 0.8131
1.7648 7.08 2050 1.2777 0.7240 0.7581 0.7240 0.7109 0.8099
1.7648 7.25 2100 1.2319 0.7288 0.7373 0.7288 0.7114 0.8123
1.7648 7.42 2150 1.2074 0.7433 0.7717 0.7433 0.7317 0.8215
1.7648 7.59 2200 1.2150 0.7433 0.7850 0.7433 0.7348 0.8235
1.7648 7.77 2250 1.1787 0.7603 0.7930 0.7603 0.7462 0.8344
1.7648 7.94 2300 1.1815 0.7676 0.7932 0.7676 0.7576 0.8404
1.7648 8.11 2350 1.1578 0.7676 0.7972 0.7676 0.7601 0.8404
1.7648 8.28 2400 1.1605 0.7651 0.7982 0.7651 0.7560 0.8387
1.7648 8.46 2450 1.1563 0.7627 0.7937 0.7627 0.7548 0.8370
1.5781 8.63 2500 1.1303 0.7579 0.7847 0.7579 0.7476 0.8337
1.5781 8.8 2550 1.1217 0.7797 0.8117 0.7797 0.7702 0.8489
1.5781 8.97 2600 1.1278 0.7724 0.8025 0.7724 0.7640 0.8438
1.5781 9.15 2650 1.1188 0.7748 0.8022 0.7748 0.7653 0.8455
1.5781 9.32 2700 1.1161 0.7676 0.7979 0.7676 0.7588 0.8404
1.5781 9.49 2750 1.1078 0.7748 0.8012 0.7748 0.7650 0.8446
1.5781 9.66 2800 1.1104 0.7724 0.7973 0.7724 0.7632 0.8429
1.5781 9.84 2850 1.1058 0.7748 0.8018 0.7748 0.7651 0.8455

Framework versions

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