File size: 2,156 Bytes
88ba474 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 |
---
base_model: ys7yoo/nli_klue_roberta_large_ep9
tags:
- generated_from_trainer
datasets:
- klue
model-index:
- name: sts_ys7yoo_nli_klue_roberta_large_ep9_ep9
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. -->
# sts_ys7yoo_nli_klue_roberta_large_ep9_ep9
This model is a fine-tuned version of [ys7yoo/nli_klue_roberta_large_ep9](https://huggingface.co/ys7yoo/nli_klue_roberta_large_ep9) on the klue dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3322
- Mse: 0.3322
- Mae: 0.4242
- R2: 0.8479
## 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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 9
### Training results
| Training Loss | Epoch | Step | Validation Loss | Mse | Mae | R2 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|
| 1.1881 | 1.0 | 183 | 0.4082 | 0.4082 | 0.4975 | 0.8131 |
| 0.1751 | 2.0 | 366 | 0.4353 | 0.4353 | 0.4964 | 0.8007 |
| 0.1222 | 3.0 | 549 | 0.3238 | 0.3238 | 0.4144 | 0.8517 |
| 0.0899 | 4.0 | 732 | 0.3434 | 0.3434 | 0.4482 | 0.8428 |
| 0.0659 | 5.0 | 915 | 0.3174 | 0.3174 | 0.4191 | 0.8547 |
| 0.0483 | 6.0 | 1098 | 0.3439 | 0.3439 | 0.4422 | 0.8425 |
| 0.0361 | 7.0 | 1281 | 0.3472 | 0.3472 | 0.4402 | 0.8410 |
| 0.0265 | 8.0 | 1464 | 0.3667 | 0.3667 | 0.4426 | 0.8321 |
| 0.0203 | 9.0 | 1647 | 0.3322 | 0.3322 | 0.4242 | 0.8479 |
### Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu117
- Datasets 2.13.0
- Tokenizers 0.13.3
|