File size: 2,039 Bytes
dd0187f |
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: klue/roberta-large
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: mango-16-0.00002-10-fin
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. -->
# mango-16-0.00002-10-fin
This model is a fine-tuned version of [klue/roberta-large](https://huggingface.co/klue/roberta-large) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 3.0500
- Accuracy: 0.6357
- F1: 0.6333
## 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: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
| No log | 1.0 | 466 | 2.1468 | 0.6171 | 0.6168 |
| 0.13 | 2.0 | 932 | 2.2649 | 0.6268 | 0.6204 |
| 0.137 | 3.0 | 1398 | 2.1698 | 0.6254 | 0.6221 |
| 0.1215 | 4.0 | 1864 | 2.1453 | 0.6265 | 0.6262 |
| 0.1048 | 5.0 | 2330 | 2.4639 | 0.6205 | 0.6214 |
| 0.0745 | 6.0 | 2796 | 2.7197 | 0.6341 | 0.6267 |
| 0.0524 | 7.0 | 3262 | 2.8052 | 0.6317 | 0.6283 |
| 0.0271 | 8.0 | 3728 | 2.9613 | 0.6297 | 0.6260 |
| 0.0146 | 9.0 | 4194 | 3.0469 | 0.6292 | 0.6282 |
| 0.0099 | 10.0 | 4660 | 3.0500 | 0.6357 | 0.6333 |
### Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0a0+b5021ba
- Datasets 2.6.2
- Tokenizers 0.14.1
|