Edit model card

vietnamese-administrative-map-name-normalization-0.0.1

This model is a fine-tuned version of vinai/bartpho-syllable on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0109
  • Model Preparation Time: 0.006
  • Sacrebleu: 98.1696
  • Cer: 0.0026
  • Wer: 0.0085

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: 128
  • eval_batch_size: 128
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 1024
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Model Preparation Time Sacrebleu Cer Wer
0.2144 2.0877 500 0.0313 0.006 95.4947 0.0075 0.0250
0.0539 4.1754 1000 0.0130 0.006 98.1454 0.0032 0.0101

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.0+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1
Downloads last month
1
Safetensors
Model size
396M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for minhhn/vietnamese-administrative-map-name-normalization-0.0.1

Finetuned
(30)
this model