xlm-roberta-base_kin-amh-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: 0.0297
- Spearman Corr: 0.7479
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 | 0.59 | 200 | 0.0429 | 0.6489 |
No log | 1.17 | 400 | 0.0316 | 0.6683 |
No log | 1.76 | 600 | 0.0254 | 0.6929 |
0.0392 | 2.35 | 800 | 0.0271 | 0.7095 |
0.0392 | 2.93 | 1000 | 0.0277 | 0.7171 |
0.0392 | 3.52 | 1200 | 0.0347 | 0.7062 |
0.0215 | 4.11 | 1400 | 0.0337 | 0.7350 |
0.0215 | 4.69 | 1600 | 0.0325 | 0.7377 |
0.0215 | 5.28 | 1800 | 0.0290 | 0.7384 |
0.0215 | 5.87 | 2000 | 0.0290 | 0.7244 |
0.0155 | 6.45 | 2200 | 0.0264 | 0.7494 |
0.0155 | 7.04 | 2400 | 0.0246 | 0.7479 |
0.0155 | 7.62 | 2600 | 0.0314 | 0.7349 |
0.0115 | 8.21 | 2800 | 0.0274 | 0.7477 |
0.0115 | 8.8 | 3000 | 0.0316 | 0.7362 |
0.0115 | 9.38 | 3200 | 0.0277 | 0.7482 |
0.0115 | 9.97 | 3400 | 0.0279 | 0.7447 |
0.0089 | 10.56 | 3600 | 0.0280 | 0.7343 |
0.0089 | 11.14 | 3800 | 0.0297 | 0.7479 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 4