metadata
tags:
- generated_from_trainer
model-index:
- name: REPEAT_3wangchanberta-base-att-spm-uncased
results: []
REPEAT_3wangchanberta-base-att-spm-uncased
This model is a fine-tuned version of airesearch/wangchanberta-base-att-spm-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.2267
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: 8
- eval_batch_size: 8
- seed: 42
- 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 |
|---|---|---|---|
| 0.4146 | 1.0 | 8561 | 0.6854 |
| 0.1716 | 2.0 | 17122 | 1.2020 |
| 0.1353 | 3.0 | 25683 | 2.0329 |
| 0.115 | 4.0 | 34244 | 2.4918 |
| 0.1031 | 5.0 | 42805 | 2.2267 |
Framework versions
- Transformers 4.15.0
- Pytorch 1.13.0+cu116
- Datasets 1.17.0
- Tokenizers 0.10.3