metadata
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
- generated_from_trainer
model-index:
- name: xlm-roberta-base_lr0.0001_seed42_basic_original_kin-hau-eng_train
results: []
xlm-roberta-base_lr0.0001_seed42_basic_original_kin-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: 0.0490
- Spearman Corr: 0.0413
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: 0.0001
- 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 | 1.59 | 200 | 0.0521 | 0.0335 |
| 0.0641 | 3.19 | 400 | 0.0491 | 0.0517 |
| 0.0587 | 4.78 | 600 | 0.0499 | 0.0819 |
| 0.0574 | 6.37 | 800 | 0.0551 | -0.0486 |
| 0.0575 | 7.97 | 1000 | 0.0498 | 0.0406 |
| 0.0575 | 9.56 | 1200 | 0.0525 | -0.0395 |
| 0.0566 | 11.16 | 1400 | 0.0499 | 0.0551 |
| 0.0564 | 12.75 | 1600 | 0.0509 | 0.0279 |
| 0.0557 | 14.34 | 1800 | 0.0513 | -0.0191 |
| 0.0558 | 15.94 | 2000 | 0.0516 | 0.0619 |
| 0.0558 | 17.53 | 2200 | 0.0511 | 0.0308 |
| 0.0555 | 19.12 | 2400 | 0.0490 | 0.0413 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2