metadata
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
- generated_from_trainer
model-index:
- name: xlm-roberta-base_lr2e-05_seed42_basic_original_kin-hau-eng_train
results: []
xlm-roberta-base_lr2e-05_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: 0.0348
- Spearman Corr: 0.7245
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 | 1.59 | 200 | 0.0325 | 0.6966 |
| 0.0428 | 3.19 | 400 | 0.0319 | 0.7262 |
| 0.0237 | 4.78 | 600 | 0.0334 | 0.7202 |
| 0.0178 | 6.37 | 800 | 0.0315 | 0.7350 |
| 0.0138 | 7.97 | 1000 | 0.0339 | 0.7365 |
| 0.0138 | 9.56 | 1200 | 0.0325 | 0.7316 |
| 0.0107 | 11.16 | 1400 | 0.0354 | 0.7334 |
| 0.0089 | 12.75 | 1600 | 0.0345 | 0.7268 |
| 0.0075 | 14.34 | 1800 | 0.0346 | 0.7321 |
| 0.0063 | 15.94 | 2000 | 0.0360 | 0.7278 |
| 0.0063 | 17.53 | 2200 | 0.0384 | 0.7253 |
| 0.0057 | 19.12 | 2400 | 0.0320 | 0.7226 |
| 0.005 | 20.72 | 2600 | 0.0337 | 0.7213 |
| 0.0047 | 22.31 | 2800 | 0.0348 | 0.7245 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2