XLM-RoBERTa-xtreme-en-token-drift
This model is a fine-tuned version of xlm-roberta-base on the xtreme_en_token_drift dataset. It achieves the following results on the evaluation set:
- Loss: 0.2802
- Accuracy: 0.9089
- F1: 0.7613
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: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.6398 | 1.0 | 161 | 0.3421 | 0.8973 | 0.7111 |
| 0.3268 | 2.0 | 322 | 0.2846 | 0.9097 | 0.7611 |
| 0.2701 | 3.0 | 483 | 0.2802 | 0.9089 | 0.7613 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1