metadata
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
- generated_from_trainer
model-index:
- name: xlm-roberta-base_lr2e-05_seed42_basic_original_esp-hau-eng_train
results: []
xlm-roberta-base_lr2e-05_seed42_basic_original_esp-hau-eng_train
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: 0.0304
- Spearman Corr: 0.7478
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 | 1.45 | 200 | 0.0245 | 0.7170 |
0.0409 | 2.91 | 400 | 0.0266 | 0.7475 |
0.0242 | 4.36 | 600 | 0.0272 | 0.7548 |
0.0242 | 5.82 | 800 | 0.0273 | 0.7521 |
0.0183 | 7.27 | 1000 | 0.0274 | 0.7521 |
0.0142 | 8.73 | 1200 | 0.0282 | 0.7645 |
0.0111 | 10.18 | 1400 | 0.0263 | 0.7587 |
0.0111 | 11.64 | 1600 | 0.0302 | 0.7507 |
0.0093 | 13.09 | 1800 | 0.0268 | 0.7509 |
0.0076 | 14.55 | 2000 | 0.0306 | 0.7383 |
0.0062 | 16.0 | 2200 | 0.0304 | 0.7478 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2