Hugging Face's logo

fydhfzh
/
hubert-classifier-aug-fold-0-old

Audio Classification
Transformers
TensorBoard
Safetensors
hubert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
hubert-classifier-aug-fold-0-old / README.md
fydhfzh's picture
fydhfzh
End of training
42c5c7d verified
preview code
|
raw
history blame
No virus
7.8 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-0
    results: []

hubert-classifier-aug-fold-0

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: 0.6445
  • Accuracy: 0.7978
  • Precision: 0.8206
  • Recall: 0.7978
  • F1: 0.7858
  • Binary: 0.8580

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 4.3930 0.0216 0.0009 0.0216 0.0017 0.1358
No log 0.38 100 4.0474 0.0270 0.0008 0.0270 0.0016 0.2286
No log 0.58 150 3.8906 0.0296 0.0013 0.0296 0.0025 0.2442
No log 0.77 200 3.6015 0.0485 0.0029 0.0485 0.0053 0.3294
No log 0.96 250 3.5656 0.0539 0.0138 0.0539 0.0140 0.3226
4.0634 1.15 300 3.4188 0.0485 0.0066 0.0485 0.0105 0.3310
4.0634 1.34 350 3.3200 0.0728 0.0113 0.0728 0.0190 0.3461
4.0634 1.53 400 3.1522 0.0809 0.0132 0.0809 0.0210 0.3534
4.0634 1.73 450 3.0267 0.1078 0.0260 0.1078 0.0363 0.3747
4.0634 1.92 500 2.9975 0.1024 0.0334 0.1024 0.0409 0.3677
3.3451 2.11 550 2.7835 0.1617 0.0980 0.1617 0.0936 0.4108
3.3451 2.3 600 2.6777 0.2237 0.1048 0.2237 0.1246 0.4561
3.3451 2.49 650 2.4946 0.2642 0.1847 0.2642 0.1754 0.4833
3.3451 2.68 700 2.3266 0.2776 0.2121 0.2776 0.1948 0.4919
3.3451 2.88 750 2.1236 0.3558 0.2876 0.3558 0.2704 0.5461
2.8034 3.07 800 1.9687 0.3908 0.3357 0.3908 0.3092 0.5712
2.8034 3.26 850 1.8541 0.4151 0.3439 0.4151 0.3410 0.5873
2.8034 3.45 900 1.7298 0.4798 0.4299 0.4798 0.4120 0.6342
2.8034 3.64 950 1.6094 0.5148 0.4804 0.5148 0.4505 0.6590
2.8034 3.84 1000 1.5433 0.5526 0.5102 0.5526 0.4910 0.6863
2.2398 4.03 1050 1.4371 0.5687 0.5274 0.5687 0.5078 0.6960
2.2398 4.22 1100 1.2423 0.6334 0.6342 0.6334 0.6024 0.7445
2.2398 4.41 1150 1.1731 0.6631 0.6428 0.6631 0.6210 0.7625
2.2398 4.6 1200 1.1174 0.7008 0.7177 0.7008 0.6778 0.7900
2.2398 4.79 1250 1.0677 0.6954 0.7081 0.6954 0.6674 0.7854
2.2398 4.99 1300 1.0534 0.6739 0.6889 0.6739 0.6480 0.7712
1.8823 5.18 1350 1.0200 0.7035 0.7035 0.7035 0.6711 0.7930
1.8823 5.37 1400 0.9667 0.7035 0.7223 0.7035 0.6864 0.7919
1.8823 5.56 1450 0.9057 0.7197 0.7433 0.7197 0.6972 0.8043
1.8823 5.75 1500 0.8284 0.7547 0.7680 0.7547 0.7348 0.8296
1.8823 5.94 1550 0.8156 0.7439 0.7708 0.7439 0.7310 0.8205
1.6355 6.14 1600 0.8034 0.7412 0.7776 0.7412 0.7313 0.8194
1.6355 6.33 1650 0.8032 0.7547 0.7768 0.7547 0.7430 0.8307
1.6355 6.52 1700 0.8030 0.7412 0.7495 0.7412 0.7195 0.8213
1.6355 6.71 1750 0.7365 0.7898 0.8217 0.7898 0.7786 0.8542
1.6355 6.9 1800 0.7149 0.7817 0.8157 0.7817 0.7653 0.8493
1.4501 7.09 1850 0.7493 0.7790 0.8200 0.7790 0.7698 0.8453
1.4501 7.29 1900 0.7022 0.8086 0.8316 0.8086 0.8003 0.8663
1.4501 7.48 1950 0.6849 0.7978 0.8226 0.7978 0.7874 0.8596
1.4501 7.67 2000 0.6008 0.8464 0.8689 0.8464 0.8436 0.8927
1.4501 7.86 2050 0.6288 0.8194 0.8449 0.8194 0.8112 0.8757
1.3207 8.05 2100 0.6410 0.8167 0.8407 0.8167 0.8051 0.8722
1.3207 8.25 2150 0.6128 0.8086 0.8318 0.8086 0.8031 0.8663
1.3207 8.44 2200 0.6072 0.8113 0.8337 0.8113 0.8053 0.8682
1.3207 8.63 2250 0.5702 0.8275 0.8519 0.8275 0.8210 0.8795
1.3207 8.82 2300 0.6059 0.8140 0.8373 0.8140 0.8079 0.8714
1.2239 9.01 2350 0.5053 0.8491 0.8685 0.8491 0.8453 0.8954
1.2239 9.2 2400 0.5551 0.8383 0.8657 0.8383 0.8339 0.8871
1.2239 9.4 2450 0.5767 0.8248 0.8571 0.8248 0.8201 0.8776
1.2239 9.59 2500 0.5514 0.8383 0.8618 0.8383 0.8331 0.8852
1.2239 9.78 2550 0.5911 0.8248 0.8542 0.8248 0.8195 0.8765
1.2239 9.97 2600 0.5498 0.8302 0.8677 0.8302 0.8267 0.8814
1.1305 10.16 2650 0.4722 0.8706 0.8928 0.8706 0.8672 0.9086
1.1305 10.35 2700 0.5509 0.8302 0.8654 0.8302 0.8261 0.8814
1.1305 10.55 2750 0.5590 0.8383 0.8713 0.8383 0.8336 0.8863
1.1305 10.74 2800 0.5053 0.8464 0.8702 0.8464 0.8424 0.8919
1.1305 10.93 2850 0.5135 0.8329 0.8548 0.8329 0.8272 0.8822
1.0868 11.12 2900 0.5438 0.8221 0.8579 0.8221 0.8145 0.8757
1.0868 11.31 2950 0.5633 0.8410 0.8720 0.8410 0.8348 0.8881
1.0868 11.51 3000 0.5050 0.8410 0.8664 0.8410 0.8389 0.8889

Framework versions

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