model_dialect / README.md
Bgeorge's picture
End of training
7df0141 verified
|
raw
history blame
2.51 kB
metadata
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: model_dialect
    results: []

model_dialect

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: 0.9645
  • Accuracy: 0.6605

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 16

Training results

Training Loss Epoch Step Validation Loss Accuracy
6.4375 0.9455 13 1.5888 0.2864
6.3195 1.9636 27 1.4956 0.3303
5.5946 2.9818 41 1.3432 0.4296
5.1778 4.0 55 1.2428 0.4896
4.8878 4.9455 68 1.1755 0.5219
4.4526 5.9636 82 1.1291 0.5612
4.2637 6.9818 96 1.1010 0.5843
3.92 8.0 110 1.0861 0.5889
3.7452 8.9455 123 1.0428 0.6051
3.6307 9.9636 137 1.0306 0.6351
3.4254 10.9818 151 0.9944 0.6467
3.1453 12.0 165 0.9931 0.6328
3.2321 12.9455 178 0.9595 0.6582
3.0323 13.9636 192 0.9593 0.6490
2.9844 14.9818 206 0.9645 0.6605
2.9844 15.1273 208 0.9652 0.6605

Framework versions

  • Transformers 4.46.0
  • Pytorch 2.4.0
  • Datasets 3.0.1
  • Tokenizers 0.20.0