bart-noised-with-kaggle-gcd-dist
This model is a fine-tuned version of gayanin/bart-noised-with-kaggle-dist on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4538
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.6061 | 0.11 | 500 | 0.5365 |
0.5537 | 0.21 | 1000 | 0.5251 |
0.5591 | 0.32 | 1500 | 0.5202 |
0.5669 | 0.43 | 2000 | 0.5069 |
0.4669 | 0.54 | 2500 | 0.5038 |
0.5457 | 0.64 | 3000 | 0.4923 |
0.5237 | 0.75 | 3500 | 0.4922 |
0.5186 | 0.86 | 4000 | 0.4802 |
0.5148 | 0.96 | 4500 | 0.4777 |
0.4127 | 1.07 | 5000 | 0.4822 |
0.4207 | 1.18 | 5500 | 0.4807 |
0.4362 | 1.28 | 6000 | 0.4770 |
0.4072 | 1.39 | 6500 | 0.4763 |
0.4503 | 1.5 | 7000 | 0.4701 |
0.3683 | 1.61 | 7500 | 0.4693 |
0.3897 | 1.71 | 8000 | 0.4636 |
0.4421 | 1.82 | 8500 | 0.4561 |
0.3836 | 1.93 | 9000 | 0.4588 |
0.3405 | 2.03 | 9500 | 0.4634 |
0.3147 | 2.14 | 10000 | 0.4682 |
0.3115 | 2.25 | 10500 | 0.4622 |
0.3153 | 2.35 | 11000 | 0.4625 |
0.3295 | 2.46 | 11500 | 0.4597 |
0.3529 | 2.57 | 12000 | 0.4564 |
0.3191 | 2.68 | 12500 | 0.4555 |
0.2974 | 2.78 | 13000 | 0.4547 |
0.3253 | 2.89 | 13500 | 0.4534 |
0.3627 | 3.0 | 14000 | 0.4538 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1
- Downloads last month
- 0