xlm-roberta-base_amh_loss_2e-05
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: 0.0248
- Spearman Corr: 0.7670
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 | 0.9 | 200 | 0.0231 | 0.7700 |
No log | 1.81 | 400 | 0.0253 | 0.7722 |
0.0023 | 2.71 | 600 | 0.0308 | 0.7629 |
0.0023 | 3.62 | 800 | 0.0260 | 0.7676 |
0.0024 | 4.52 | 1000 | 0.0238 | 0.7693 |
0.0024 | 5.43 | 1200 | 0.0236 | 0.7646 |
0.0019 | 6.33 | 1400 | 0.0240 | 0.7648 |
0.0019 | 7.24 | 1600 | 0.0241 | 0.7668 |
0.0018 | 8.14 | 1800 | 0.0248 | 0.7670 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 7