<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
tiny-mlm-snli-from-scratch-target-conll2003
This model is a fine-tuned version of muhtasham/tiny-mlm-snli-from-scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3282
- Precision: 0.4518
- Recall: 0.5961
- F1: 0.5140
- Accuracy: 0.9043
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 |
---|---|---|---|---|---|---|---|
0.9955 | 1.14 | 500 | 0.7185 | 0.2123 | 0.1277 | 0.1595 | 0.8166 |
0.6054 | 2.28 | 1000 | 0.5641 | 0.2535 | 0.2804 | 0.2663 | 0.8481 |
0.5153 | 3.42 | 1500 | 0.5015 | 0.2936 | 0.3475 | 0.3183 | 0.8581 |
0.4561 | 4.56 | 2000 | 0.4556 | 0.3416 | 0.4127 | 0.3738 | 0.8694 |
0.4066 | 5.69 | 2500 | 0.4141 | 0.3547 | 0.4801 | 0.4080 | 0.8784 |
0.3671 | 6.83 | 3000 | 0.3833 | 0.3941 | 0.5157 | 0.4467 | 0.8888 |
0.3301 | 7.97 | 3500 | 0.3623 | 0.4062 | 0.5559 | 0.4694 | 0.8936 |
0.2999 | 9.11 | 4000 | 0.3464 | 0.4132 | 0.5762 | 0.4813 | 0.8976 |
0.2747 | 10.25 | 4500 | 0.3369 | 0.4332 | 0.5816 | 0.4966 | 0.9012 |
0.2582 | 11.39 | 5000 | 0.3282 | 0.4518 | 0.5961 | 0.5140 | 0.9043 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2