<!-- 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. -->
small-mlm-snli-from-scratch-target-conll2003
This model is a fine-tuned version of muhtasham/small-mlm-snli-from-scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2722
- Precision: 0.6350
- Recall: 0.7334
- F1: 0.6807
- Accuracy: 0.9355
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 |
---|---|---|---|---|---|---|---|
0.5969 | 1.14 | 500 | 0.4174 | 0.3902 | 0.4960 | 0.4368 | 0.8740 |
0.3356 | 2.28 | 1000 | 0.3105 | 0.4951 | 0.6070 | 0.5454 | 0.9055 |
0.2397 | 3.42 | 1500 | 0.2794 | 0.4914 | 0.6572 | 0.5624 | 0.9120 |
0.1822 | 4.56 | 2000 | 0.2548 | 0.5653 | 0.6718 | 0.6140 | 0.9241 |
0.1443 | 5.69 | 2500 | 0.2484 | 0.5930 | 0.6816 | 0.6342 | 0.9280 |
0.1151 | 6.83 | 3000 | 0.2457 | 0.5781 | 0.7060 | 0.6357 | 0.9291 |
0.0927 | 7.97 | 3500 | 0.2430 | 0.6057 | 0.7174 | 0.6569 | 0.9320 |
0.0739 | 9.11 | 4000 | 0.2645 | 0.6313 | 0.7078 | 0.6674 | 0.9349 |
0.0606 | 10.25 | 4500 | 0.2651 | 0.6354 | 0.7218 | 0.6759 | 0.9359 |
0.0506 | 11.39 | 5000 | 0.2722 | 0.6350 | 0.7334 | 0.6807 | 0.9355 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2