metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
datasets:
- wmt20_mlqe_task1
model-index:
- name: xlmr-ne-en-no_shuffled-orig-test1000
results: []
xlmr-ne-en-no_shuffled-orig-test1000
This model is a fine-tuned version of xlm-roberta-base on the wmt20_mlqe_task1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5471
- R Squared: 0.2194
- Mae: 0.5682
- Pearson R: 0.6846
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | R Squared | Mae | Pearson R |
|---|---|---|---|---|---|---|
| No log | 1.0 | 438 | 0.5304 | 0.2432 | 0.5810 | 0.6277 |
| 0.7677 | 2.0 | 876 | 0.4347 | 0.3798 | 0.5171 | 0.6653 |
| 0.5573 | 3.0 | 1314 | 0.5234 | 0.2531 | 0.5506 | 0.6777 |
| 0.4018 | 4.0 | 1752 | 0.5984 | 0.1462 | 0.5975 | 0.6819 |
| 0.2964 | 5.0 | 2190 | 0.5471 | 0.2194 | 0.5682 | 0.6846 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.6
- Tokenizers 0.14.1