20NG_ELECTRA_5E
This model is a fine-tuned version of google/electra-small-discriminator on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.8534
- Accuracy: 0.1733
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.9938 | 0.07 | 50 | 2.9881 | 0.1067 |
| 2.9879 | 0.14 | 100 | 2.9852 | 0.1133 |
| 2.9842 | 0.21 | 150 | 2.9796 | 0.1333 |
| 2.9722 | 0.28 | 200 | 2.9653 | 0.1733 |
| 2.9559 | 0.35 | 250 | 2.9458 | 0.18 |
| 2.9368 | 0.42 | 300 | 2.9257 | 0.1667 |
| 2.9234 | 0.49 | 350 | 2.9101 | 0.18 |
| 2.9094 | 0.56 | 400 | 2.8926 | 0.1733 |
| 2.898 | 0.64 | 450 | 2.8810 | 0.18 |
| 2.8941 | 0.71 | 500 | 2.8715 | 0.18 |
| 2.8742 | 0.78 | 550 | 2.8646 | 0.1533 |
| 2.8765 | 0.85 | 600 | 2.8597 | 0.1867 |
| 2.8738 | 0.92 | 650 | 2.8555 | 0.1733 |
| 2.866 | 0.99 | 700 | 2.8534 | 0.1733 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1
- Datasets 2.8.0
- Tokenizers 0.13.2