xlm-roberta-base_latin_original_amh-hau-eng_train_spearman_corr
This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0279
- Spearman Corr: 0.7633
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: 128
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Spearman Corr |
---|---|---|---|---|
No log | 1.55 | 200 | 0.0217 | 0.7599 |
0.049 | 3.1 | 400 | 0.0217 | 0.7806 |
0.0235 | 4.65 | 600 | 0.0270 | 0.7760 |
0.0177 | 6.2 | 800 | 0.0228 | 0.7779 |
0.0177 | 7.75 | 1000 | 0.0230 | 0.7721 |
0.0141 | 9.3 | 1200 | 0.0260 | 0.7493 |
0.0112 | 10.85 | 1400 | 0.0292 | 0.7630 |
0.009 | 12.4 | 1600 | 0.0266 | 0.7656 |
0.0075 | 13.95 | 1800 | 0.0263 | 0.7696 |
0.0075 | 15.5 | 2000 | 0.0279 | 0.7633 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 7