metadata
license: mit
base_model: yomilimi/Gyeongsang_decoder
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: Jeolla_encoder
results: []
Jeolla_encoder
This model is a fine-tuned version of yomilimi/Gyeongsang_decoder on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0126
- Bleu: 89.0253
- Gen Len: 14.0976
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 |
|---|---|---|---|---|---|
| 0.0152 | 1.0 | 15477 | 0.0140 | 88.7453 | 14.1001 |
| 0.0133 | 2.0 | 30954 | 0.0130 | 88.9598 | 14.0964 |
| 0.0109 | 3.0 | 46431 | 0.0126 | 89.0253 | 14.0976 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1