|
--- |
|
base_model: klue/roberta-large |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: reward-bert-duplicate-answer-300 |
|
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. --> |
|
|
|
# reward-bert-duplicate-answer-300 |
|
|
|
This model is a fine-tuned version of [klue/roberta-large](https://huggingface.co/klue/roberta-large) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2419 |
|
- Accuracy: 0.0 |
|
|
|
## 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: 9e-05 |
|
- train_batch_size: 6 |
|
- eval_batch_size: 6 |
|
- seed: 2023 |
|
- gradient_accumulation_steps: 10 |
|
- total_train_batch_size: 60 |
|
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 0.5015 | 0.17 | 100 | 0.5284 | 0.0 | |
|
| 0.4259 | 0.34 | 200 | 0.3848 | 0.0 | |
|
| 0.3808 | 0.51 | 300 | 0.2962 | 0.0 | |
|
| 0.3328 | 0.69 | 400 | 0.2592 | 0.0 | |
|
| 0.2086 | 0.86 | 500 | 0.2419 | 0.0 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.1+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
