tiny-mlm-wikitext-target-conll2003
This model is a fine-tuned version of muhtasham/tiny-mlm-wikitext on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1476
- Precision: 0.7356
- Recall: 0.8095
- F1: 0.7708
- Accuracy: 0.9566
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.7563 | 1.14 | 500 | 0.4302 | 0.4626 | 0.5145 | 0.4872 | 0.8907 |
| 0.3735 | 2.28 | 1000 | 0.2915 | 0.5885 | 0.6547 | 0.6198 | 0.9234 |
| 0.2875 | 3.42 | 1500 | 0.2354 | 0.6404 | 0.7070 | 0.6721 | 0.9342 |
| 0.2366 | 4.56 | 2000 | 0.2071 | 0.6771 | 0.7408 | 0.7075 | 0.9411 |
| 0.2038 | 5.69 | 2500 | 0.1847 | 0.6981 | 0.7657 | 0.7303 | 0.9465 |
| 0.18 | 6.83 | 3000 | 0.1711 | 0.7112 | 0.7841 | 0.7459 | 0.9500 |
| 0.159 | 7.97 | 3500 | 0.1626 | 0.7125 | 0.7908 | 0.7496 | 0.9524 |
| 0.1415 | 9.11 | 4000 | 0.1554 | 0.7193 | 0.7933 | 0.7545 | 0.9544 |
| 0.1276 | 10.25 | 4500 | 0.1514 | 0.7366 | 0.8029 | 0.7683 | 0.9558 |
| 0.1184 | 11.39 | 5000 | 0.1476 | 0.7356 | 0.8095 | 0.7708 | 0.9566 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2