metadata
tags:
- generated_from_trainer
base_model: airesearch/wangchanberta-base-att-spm-uncased
metrics:
- accuracy
model-index:
- name: fine-tune-wangchanberta-AJA-split-headline1
results: []
fine-tune-wangchanberta-AJA-split-headline1
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: 1.1001
- Accuracy: 0.3960
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: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.1119 | 1.0 | 26 | 1.0998 | 0.3713 |
| 1.1023 | 2.0 | 52 | 1.1027 | 0.3960 |
| 1.085 | 3.0 | 78 | 1.1007 | 0.3267 |
| 1.0799 | 4.0 | 104 | 1.1020 | 0.4010 |
| 1.0779 | 5.0 | 130 | 1.0973 | 0.4010 |
| 1.0668 | 6.0 | 156 | 1.1069 | 0.3218 |
| 1.0648 | 7.0 | 182 | 1.1001 | 0.3960 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1