You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

WMACv2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2606
  • Wer: 41.3106

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: 2.5e-05
  • train_batch_size: 8
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 650
  • training_steps: 5000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
1.5465 0.4757 1000 0.4199 54.1280
1.0508 0.9514 2000 0.2911 43.9422
0.6641 1.4267 3000 0.2649 44.7884
0.4569 1.9024 4000 0.2369 41.4448
0.2753 2.3777 5000 0.2606 41.3106

Framework versions

  • Transformers 5.2.0
  • Pytorch 2.10.0+cu128
  • Datasets 3.6.0
  • Tokenizers 0.22.2
Downloads last month
111
Safetensors
Model size
0.7B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for 7ocho/WMACv2

Finetunes
1 model