xlm-roberta-base_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: 0.0236
- Spearman Corr: 0.7453
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 | 0.0398 | 0.7151 |
No log | 1.06 | 400 | 0.0250 | 0.7514 |
No log | 1.59 | 600 | 0.0180 | 0.7542 |
0.0322 | 2.12 | 800 | 0.0223 | 0.7529 |
0.0322 | 2.65 | 1000 | 0.0250 | 0.7695 |
0.0322 | 3.17 | 1200 | 0.0238 | 0.7662 |
0.0322 | 3.7 | 1400 | 0.0226 | 0.7696 |
0.0174 | 4.23 | 1600 | 0.0227 | 0.7595 |
0.0174 | 4.76 | 1800 | 0.0277 | 0.7616 |
0.0174 | 5.29 | 2000 | 0.0244 | 0.7598 |
0.0174 | 5.82 | 2200 | 0.0236 | 0.7515 |
0.0118 | 6.35 | 2400 | 0.0235 | 0.7612 |
0.0118 | 6.88 | 2600 | 0.0226 | 0.7398 |
0.0118 | 7.41 | 2800 | 0.0236 | 0.7471 |
0.0118 | 7.94 | 3000 | 0.0236 | 0.7453 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 0