xlm-roberta-base_latin_amh-esp-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: 0.0232
- Spearman Corr: 0.7591
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 | 0.53 | 200 | 0.0205 | 0.7282 |
No log | 1.06 | 400 | 0.0229 | 0.7315 |
No log | 1.59 | 600 | 0.0210 | 0.7374 |
0.0323 | 2.12 | 800 | 0.0250 | 0.7461 |
0.0323 | 2.65 | 1000 | 0.0209 | 0.7509 |
0.0323 | 3.17 | 1200 | 0.0258 | 0.7498 |
0.0323 | 3.7 | 1400 | 0.0185 | 0.7560 |
0.0197 | 4.23 | 1600 | 0.0207 | 0.7505 |
0.0197 | 4.76 | 1800 | 0.0212 | 0.7509 |
0.0197 | 5.29 | 2000 | 0.0222 | 0.7641 |
0.0197 | 5.82 | 2200 | 0.0258 | 0.7585 |
0.014 | 6.35 | 2400 | 0.0236 | 0.7655 |
0.014 | 6.88 | 2600 | 0.0254 | 0.7625 |
0.014 | 7.41 | 2800 | 0.0283 | 0.7504 |
0.014 | 7.94 | 3000 | 0.0306 | 0.7585 |
0.0107 | 8.47 | 3200 | 0.0236 | 0.7519 |
0.0107 | 8.99 | 3400 | 0.0249 | 0.7506 |
0.0107 | 9.52 | 3600 | 0.0235 | 0.7487 |
0.008 | 10.05 | 3800 | 0.0256 | 0.7613 |
0.008 | 10.58 | 4000 | 0.0232 | 0.7591 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 6