NMP123's picture
End of training
d77a1c0 verified
|
raw
history blame
2.08 kB
metadata
library_name: transformers
license: mit
base_model: facebook/w2v-bert-2.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_17_0
metrics:
  - wer
model-index:
  - name: w2v-bert-2.0-Vietnamese-colab-CV17.0
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_17_0
          type: common_voice_17_0
          config: vi
          split: test
          args: vi
        metrics:
          - name: Wer
            type: wer
            value: 0.26461245235069886

w2v-bert-2.0-Vietnamese-colab-CV17.0

This model is a fine-tuned version of facebook/w2v-bert-2.0 on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7535
  • Wer: 0.2646

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: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Use OptimizerNames.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_steps: 500
  • num_epochs: 10
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
3.2974 3.2609 300 0.6899 0.3559
0.1432 6.5217 600 0.7335 0.2991
0.0363 9.7826 900 0.7535 0.2646

Framework versions

  • Transformers 4.50.0
  • Pytorch 2.6.0+cu124
  • Datasets 3.4.1
  • Tokenizers 0.21.1