head_l23_const_lr_1e-4

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:

  • eval_loss: 1.6613
  • eval_wer: 0.9996
  • eval_cer: 0.5802
  • eval_runtime: 227.132
  • eval_samples_per_second: 18.848
  • eval_steps_per_second: 2.36
  • epoch: 63.0769
  • step: 20500

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: 16
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • num_epochs: 1000
  • mixed_precision_training: Native AMP

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1
Downloads last month
14
Safetensors
Model size
581M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Model tree for LevonHakobyan/head_l23_const_lr_1e-4

Finetuned
(252)
this model