small-mlm-tweet_eval-from-scratch-custom-tokenizer-target-conll2003
This model is a fine-tuned version of muhtasham/small-mlm-tweet_eval-from-scratch-custom-tokenizer on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3269
- Precision: 0.4715
- Recall: 0.6299
- F1: 0.5393
- Accuracy: 0.9048
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 |
|---|---|---|---|---|---|---|---|
| 0.6782 | 1.14 | 500 | 0.5064 | 0.2100 | 0.3012 | 0.2475 | 0.8398 |
| 0.4636 | 2.28 | 1000 | 0.4216 | 0.2659 | 0.4042 | 0.3208 | 0.8626 |
| 0.3915 | 3.42 | 1500 | 0.3899 | 0.2693 | 0.4608 | 0.3399 | 0.8674 |
| 0.3363 | 4.56 | 2000 | 0.3541 | 0.3403 | 0.4768 | 0.3972 | 0.8820 |
| 0.2933 | 5.69 | 2500 | 0.3351 | 0.3719 | 0.5355 | 0.4389 | 0.8879 |
| 0.2551 | 6.83 | 3000 | 0.3247 | 0.4078 | 0.5670 | 0.4744 | 0.8956 |
| 0.2196 | 7.97 | 3500 | 0.3208 | 0.4562 | 0.5602 | 0.5029 | 0.9002 |
| 0.1871 | 9.11 | 4000 | 0.3226 | 0.4834 | 0.5865 | 0.5300 | 0.9045 |
| 0.1575 | 10.25 | 4500 | 0.3180 | 0.4671 | 0.6195 | 0.5326 | 0.9048 |
| 0.1366 | 11.39 | 5000 | 0.3269 | 0.4715 | 0.6299 | 0.5393 | 0.9048 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2