job-listing-relevance-model
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1649
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7435 | 0.43 | 50 | 0.6889 |
| 0.3222 | 0.87 | 100 | 0.2906 |
| 0.2573 | 1.3 | 150 | 0.1937 |
| 0.1205 | 1.74 | 200 | 0.1411 |
| 0.1586 | 2.17 | 250 | 0.2008 |
| 0.0755 | 2.61 | 300 | 0.1926 |
| 0.062 | 3.04 | 350 | 0.2257 |
| 0.0644 | 3.48 | 400 | 0.1497 |
| 0.1034 | 3.91 | 450 | 0.1561 |
| 0.008 | 4.35 | 500 | 0.2067 |
| 0.0616 | 4.78 | 550 | 0.2067 |
| 0.0766 | 5.22 | 600 | 0.1494 |
| 0.0029 | 5.65 | 650 | 0.2078 |
| 0.1076 | 6.09 | 700 | 0.1669 |
| 0.0025 | 6.52 | 750 | 0.1564 |
| 0.0498 | 6.95 | 800 | 0.2355 |
| 0.0011 | 7.39 | 850 | 0.1652 |
| 0.0271 | 7.82 | 900 | 0.1731 |
| 0.012 | 8.26 | 950 | 0.1590 |
| 0.0257 | 8.69 | 1000 | 0.1638 |
| 0.0009 | 9.13 | 1050 | 0.1851 |
| 0.0013 | 9.56 | 1100 | 0.1613 |
| 0.0015 | 10.0 | 1150 | 0.1649 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.11.0+cu113
- Datasets 2.0.0
- Tokenizers 0.11.6