|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: kobigbird-bert-base-finetuned-klue |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# kobigbird-bert-base-finetuned-klue |
|
|
|
This model is a fine-tuned version of [monologg/kobigbird-bert-base](https://huggingface.co/monologg/kobigbird-bert-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.9124 |
|
|
|
## 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: 5e-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: 60 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 5.1951 | 0.69 | 500 | 3.9518 | |
|
| 3.7508 | 1.38 | 1000 | 3.4423 | |
|
| 3.4379 | 2.07 | 1500 | 3.1397 | |
|
| 3.1865 | 2.76 | 2000 | 2.8141 | |
|
| 2.9852 | 3.45 | 2500 | 2.6055 | |
|
| 2.7907 | 4.14 | 3000 | 2.4892 | |
|
| 2.6985 | 4.83 | 3500 | 2.4130 | |
|
| 2.63 | 5.52 | 4000 | 2.3479 | |
|
| 2.5616 | 6.21 | 4500 | 2.3134 | |
|
| 2.4898 | 6.9 | 5000 | 2.2526 | |
|
| 2.4167 | 7.59 | 5500 | 2.2422 | |
|
| 2.3861 | 8.28 | 6000 | 2.2138 | |
|
| 2.3401 | 8.97 | 6500 | 2.1762 | |
|
| 2.292 | 9.66 | 7000 | 2.1417 | |
|
| 2.2705 | 10.34 | 7500 | 2.1148 | |
|
| 2.2139 | 11.03 | 8000 | 2.1357 | |
|
| 2.187 | 11.72 | 8500 | 2.0995 | |
|
| 2.1518 | 12.41 | 9000 | 2.0554 | |
|
| 2.1179 | 13.1 | 9500 | 2.0467 | |
|
| 2.0781 | 13.79 | 10000 | 2.0418 | |
|
| 2.0457 | 14.48 | 10500 | 2.0057 | |
|
| 2.0326 | 15.17 | 11000 | 2.0022 | |
|
| 1.9976 | 15.86 | 11500 | 2.0137 | |
|
| 1.9602 | 16.55 | 12000 | 1.9703 | |
|
| 1.9435 | 17.24 | 12500 | 1.9920 | |
|
| 1.9092 | 17.93 | 13000 | 1.9504 | |
|
| 1.8714 | 18.62 | 13500 | 1.9731 | |
|
| 1.8715 | 19.31 | 14000 | 1.9471 | |
|
| 1.8824 | 20.0 | 14500 | 1.9462 | |
|
| 1.8261 | 20.69 | 15000 | 1.9412 | |
|
| 1.815 | 21.38 | 15500 | 1.9308 | |
|
| 1.7988 | 22.07 | 16000 | 1.9285 | |
|
| 1.7722 | 22.76 | 16500 | 1.9341 | |
|
| 1.7606 | 23.45 | 17000 | 1.9007 | |
|
| 1.7402 | 24.14 | 17500 | 1.9206 | |
|
| 1.7177 | 24.83 | 18000 | 1.9125 | |
|
| 1.7117 | 25.52 | 18500 | 1.9178 | |
|
| 1.6834 | 26.21 | 19000 | 1.9082 | |
|
| 1.6893 | 26.9 | 19500 | 1.9124 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.11.0.dev20211231+cu113 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|
