swin-gpt-NWPU-captioning
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5339
- Accuracy: 85.19
- Bleu-1: 0.8630
- Bleu-2: 0.7745
- Bleu-3: 0.7039
- Bleu-4: 0.6472
- Meteor: 0.7734
- Rouge-l: 0.7624
- Cider: 1.8392
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 50
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1024
- num_epochs: 128
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Bleu-1 | Bleu-2 | Bleu-3 | Bleu-4 | Meteor | Rouge-l | Cider |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.7902 | 1.0 | 1969 | 0.4962 | 85.07 | 0.8231 | 0.7203 | 0.6407 | 0.5778 | 0.7196 | 0.7062 | 1.6480 |
| 0.4833 | 2.0 | 3938 | 0.4726 | 84.85 | 0.8344 | 0.7345 | 0.6529 | 0.5878 | 0.7337 | 0.7202 | 1.6506 |
| 0.4490 | 3.0 | 5907 | 0.4691 | 85.1 | 0.8574 | 0.7657 | 0.6910 | 0.6307 | 0.7639 | 0.7486 | 1.7462 |
| 0.4245 | 4.0 | 7876 | 0.4678 | 84.65 | 0.8572 | 0.7685 | 0.6960 | 0.6376 | 0.7748 | 0.7565 | 1.7604 |
| 0.4068 | 5.0 | 9845 | 0.4723 | 85.14 | 0.8616 | 0.7719 | 0.6995 | 0.6414 | 0.7781 | 0.7632 | 1.8423 |
| 0.3889 | 6.0 | 11814 | 0.4778 | 85.36 | 0.8691 | 0.7861 | 0.7186 | 0.6639 | 0.7755 | 0.7698 | 1.8639 |
| 0.3711 | 7.0 | 13783 | 0.4849 | 85.0 | 0.8599 | 0.7701 | 0.6981 | 0.6398 | 0.7815 | 0.7663 | 1.8020 |
| 0.3553 | 8.0 | 15752 | 0.4919 | 85.26 | 0.8672 | 0.7788 | 0.7071 | 0.6489 | 0.7775 | 0.7667 | 1.8345 |
| 0.3396 | 9.0 | 17721 | 0.5033 | 85.1 | 0.8621 | 0.7749 | 0.7052 | 0.6490 | 0.7690 | 0.7617 | 1.8219 |
| 0.3230 | 10.0 | 19690 | 0.5192 | 85.21 | 0.8592 | 0.7720 | 0.7017 | 0.6441 | 0.7736 | 0.7631 | 1.8233 |
| 0.3088 | 11.0 | 21659 | 0.5339 | 85.19 | 0.8630 | 0.7745 | 0.7039 | 0.6472 | 0.7734 | 0.7624 | 1.8392 |
Framework versions
- Transformers 5.12.1
- Pytorch 2.12.1+cu130
- Datasets 5.0.0
- Tokenizers 0.22.2