vit-hymenoptera-ants-bees
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the uisikdag/hymenoptera-ants-bees dataset. It achieves the following results on the evaluation set:
- Loss: 0.5957
- Accuracy: 0.6993
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: 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
- num_epochs: 10.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7319 | 1.0 | 16 | 0.7800 | 0.4967 |
| 0.6671 | 2.0 | 32 | 0.7156 | 0.5556 |
| 0.6432 | 3.0 | 48 | 0.6756 | 0.6078 |
| 0.5753 | 4.0 | 64 | 0.6586 | 0.6405 |
| 0.6145 | 5.0 | 80 | 0.6264 | 0.6797 |
| 0.5756 | 6.0 | 96 | 0.5997 | 0.6536 |
| 0.5039 | 7.0 | 112 | 0.5957 | 0.6993 |
| 0.5937 | 8.0 | 128 | 0.5787 | 0.6993 |
| 0.5324 | 9.0 | 144 | 0.5797 | 0.6863 |
| 0.4880 | 10.0 | 160 | 0.5859 | 0.6993 |
Framework versions
- Transformers 5.10.2
- Pytorch 2.10.0+cu130
- Datasets 4.1.1
- Tokenizers 0.22.1