qa_kor_math / README.md
idah4's picture
Model save
a7bff75 verified
|
raw
history blame
1.62 kB
---
license: mit
base_model: gogamza/kobart-base-v2
tags:
- generated_from_trainer
model-index:
- name: qa_kor_math
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. -->
# qa_kor_math
This model is a fine-tuned version of [gogamza/kobart-base-v2](https://huggingface.co./gogamza/kobart-base-v2) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2196
## 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: 0.0001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 400
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| No log | 0.63 | 100 | 0.6907 |
| No log | 1.26 | 200 | 0.3665 |
| No log | 1.89 | 300 | 0.3132 |
| No log | 2.52 | 400 | 0.2873 |
| 1.1333 | 3.14 | 500 | 0.2600 |
| 1.1333 | 3.77 | 600 | 0.2312 |
| 1.1333 | 4.4 | 700 | 0.2196 |
### Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2