xlm-roberta-base_lr5e-06_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: 0.0334
- Spearman Corr: 0.7247
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: 5e-06
- 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 | 0.0366 | 0.6063 |
0.0656 | 3.19 | 400 | 0.0336 | 0.6631 |
0.0346 | 4.78 | 600 | 0.0303 | 0.6995 |
0.029 | 6.37 | 800 | 0.0311 | 0.7091 |
0.0253 | 7.97 | 1000 | 0.0298 | 0.7144 |
0.0253 | 9.56 | 1200 | 0.0322 | 0.7142 |
0.0222 | 11.16 | 1400 | 0.0303 | 0.7187 |
0.0202 | 12.75 | 1600 | 0.0316 | 0.7215 |
0.0183 | 14.34 | 1800 | 0.0325 | 0.7235 |
0.0174 | 15.94 | 2000 | 0.0298 | 0.7277 |
0.0174 | 17.53 | 2200 | 0.0315 | 0.7305 |
0.0162 | 19.12 | 2400 | 0.0338 | 0.7305 |
0.015 | 20.72 | 2600 | 0.0320 | 0.7264 |
0.0145 | 22.31 | 2800 | 0.0319 | 0.7278 |
0.014 | 23.9 | 3000 | 0.0334 | 0.7247 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 3
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.