<!-- 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-custom-tokenizer-target-conll2003
This model is a fine-tuned version of muhtasham/small-mlm-snli-from-scratch-custom-tokenizer on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3378
- Precision: 0.4244
- Recall: 0.5441
- F1: 0.4768
- Accuracy: 0.8956
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 |
---|---|---|---|---|---|---|---|
0.7215 | 1.14 | 500 | 0.5706 | 0.1492 | 0.2666 | 0.1913 | 0.8173 |
0.5251 | 2.28 | 1000 | 0.4653 | 0.2138 | 0.3122 | 0.2538 | 0.8465 |
0.4538 | 3.42 | 1500 | 0.4361 | 0.2118 | 0.4010 | 0.2772 | 0.8472 |
0.3962 | 4.56 | 2000 | 0.3844 | 0.2868 | 0.4187 | 0.3404 | 0.8677 |
0.356 | 5.69 | 2500 | 0.3705 | 0.3088 | 0.4731 | 0.3737 | 0.8702 |
0.3256 | 6.83 | 3000 | 0.3540 | 0.3297 | 0.4798 | 0.3908 | 0.8790 |
0.2923 | 7.97 | 3500 | 0.3442 | 0.3762 | 0.4845 | 0.4235 | 0.8851 |
0.2599 | 9.11 | 4000 | 0.3410 | 0.3992 | 0.5101 | 0.4479 | 0.8904 |
0.2345 | 10.25 | 4500 | 0.3375 | 0.4118 | 0.5279 | 0.4627 | 0.8928 |
0.2119 | 11.39 | 5000 | 0.3378 | 0.4244 | 0.5441 | 0.4768 | 0.8956 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2