xlm_70k_ko_duoi
This model is a fine-tuned version of FacebookAI/xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3916
- Accuracy: 0.9120
- F1: 0.9120
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: cosine_with_restarts
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
---|---|---|---|---|---|
0.3819 | 1.0 | 1719 | 0.2560 | 0.8848 | 0.8855 |
0.2608 | 2.0 | 3438 | 0.2473 | 0.8967 | 0.8978 |
0.2129 | 3.0 | 5157 | 0.2387 | 0.9061 | 0.9057 |
0.1776 | 4.0 | 6876 | 0.2613 | 0.9090 | 0.9101 |
0.1522 | 5.0 | 8595 | 0.2629 | 0.9097 | 0.9101 |
0.1261 | 6.0 | 10314 | 0.2986 | 0.9110 | 0.9110 |
0.108 | 7.0 | 12033 | 0.3144 | 0.9110 | 0.9109 |
0.094 | 8.0 | 13752 | 0.3806 | 0.9117 | 0.9120 |
0.0872 | 9.0 | 15471 | 0.3900 | 0.9124 | 0.9124 |
0.0796 | 10.0 | 17190 | 0.3916 | 0.9120 | 0.9120 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 5