metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: single_label_unbiased_relevant_profession
results: []
single_label_unbiased_relevant_profession
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: 1.5581
- Acc At K: 0.8934
- Acc: 0.5742
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Acc At K | Acc |
|---|---|---|---|---|---|
| 4.1013 | 0.5 | 22700 | 2.6410 | 0.7356 | 0.4504 |
| 2.3359 | 1.0 | 45400 | 2.1112 | 0.8115 | 0.4979 |
| 1.9045 | 1.5 | 68100 | 1.9027 | 0.8428 | 0.5240 |
| 1.7084 | 2.0 | 90800 | 1.7826 | 0.8607 | 0.5340 |
| 1.5155 | 2.5 | 113500 | 1.7117 | 0.8711 | 0.5444 |
| 1.4211 | 3.0 | 136200 | 1.6643 | 0.8782 | 0.5493 |
| 1.2865 | 3.5 | 158900 | 1.6342 | 0.8812 | 0.5568 |
| 1.2357 | 4.0 | 181600 | 1.6077 | 0.8852 | 0.5588 |
| 1.1303 | 4.5 | 204300 | 1.6023 | 0.8873 | 0.5632 |
| 1.0987 | 5.0 | 227000 | 1.5784 | 0.8896 | 0.5652 |
| 1.0186 | 5.5 | 249700 | 1.5782 | 0.8904 | 0.5673 |
| 0.9982 | 6.0 | 272400 | 1.5712 | 0.8914 | 0.5707 |
| 0.9404 | 6.5 | 295100 | 1.5685 | 0.8920 | 0.5710 |
| 0.9263 | 7.0 | 317800 | 1.5615 | 0.8925 | 0.5725 |
| 0.8839 | 7.5 | 340500 | 1.5603 | 0.8929 | 0.5741 |
| 0.878 | 8.0 | 363200 | 1.5581 | 0.8934 | 0.5742 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.0+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3