Edit model card

ko_gyeongsang_test

This model is a fine-tuned version of gogamza/kobart-base-v2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.7493
  • Bleu: 90.5462
  • Gen Len: 20.0

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: 2e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
3.7799 1.0 12910 3.7647 89.6987 20.0
3.7679 2.0 25820 3.7528 90.4226 20.0
3.757 3.0 38730 3.7493 90.5462 20.0

Framework versions

  • Transformers 4.42.0.dev0
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.2
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
124M 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 yomilimi/ko_gyeongsang_test

Finetuned
(13)
this model