tiny-mlm-tweet-target-conll2003
This model is a fine-tuned version of muhtasham/tiny-mlm-tweet on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1472
- Precision: 0.7358
- Recall: 0.8098
- F1: 0.7710
- Accuracy: 0.9568
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 |
|---|---|---|---|---|---|---|---|
| 0.754 | 1.14 | 500 | 0.4280 | 0.4589 | 0.5084 | 0.4824 | 0.8922 |
| 0.3717 | 2.28 | 1000 | 0.2900 | 0.5875 | 0.6503 | 0.6173 | 0.9240 |
| 0.2865 | 3.42 | 1500 | 0.2349 | 0.6354 | 0.7050 | 0.6684 | 0.9345 |
| 0.2358 | 4.56 | 2000 | 0.2070 | 0.6760 | 0.7400 | 0.7066 | 0.9410 |
| 0.2035 | 5.69 | 2500 | 0.1835 | 0.6966 | 0.7647 | 0.7291 | 0.9467 |
| 0.1793 | 6.83 | 3000 | 0.1702 | 0.7107 | 0.7819 | 0.7446 | 0.9501 |
| 0.1579 | 7.97 | 3500 | 0.1613 | 0.7162 | 0.7932 | 0.7527 | 0.9527 |
| 0.1408 | 9.11 | 4000 | 0.1545 | 0.7209 | 0.7940 | 0.7557 | 0.9545 |
| 0.1267 | 10.25 | 4500 | 0.1505 | 0.7391 | 0.8024 | 0.7695 | 0.9559 |
| 0.1179 | 11.39 | 5000 | 0.1472 | 0.7358 | 0.8098 | 0.7710 | 0.9568 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2