alirzb's picture
Model save
50284bb verified
metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: SeizureClassifier_Wav2Vec_U_43828667
    results: []

SeizureClassifier_Wav2Vec_U_43828667

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0102
  • Accuracy: 0.9984

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: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 15
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.1571 1.0 339 0.1531 0.9665
0.0698 2.0 678 0.0642 0.9810
0.0265 3.0 1017 0.0340 0.9926
0.04 4.0 1357 0.0290 0.9903
0.0028 5.0 1696 0.0285 0.9942
0.0014 6.0 2035 0.0185 0.9965
0.0009 7.0 2374 0.0281 0.9955
0.0208 8.0 2714 0.0154 0.9974
0.0006 9.0 3053 0.0205 0.9968
0.0004 10.0 3392 0.0165 0.9974
0.0003 11.0 3731 0.0124 0.9977
0.0003 12.0 4071 0.0171 0.9971
0.0095 13.0 4410 0.0140 0.9971
0.0002 14.0 4749 0.0123 0.9984
0.0002 14.99 5085 0.0102 0.9984

Framework versions

  • Transformers 4.36.2
  • Pytorch 2.1.2+cu118
  • Datasets 2.16.1
  • Tokenizers 0.15.0