convnext-tiny-upgrade-384-batch-16
This model is a fine-tuned version of facebook/convnextv2-tiny-22k-384 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2389
- Accuracy: 0.9369
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.855 | 1.0 | 1099 | 0.4926 | 0.8577 |
| 0.6243 | 2.0 | 2198 | 0.3875 | 0.8911 |
| 0.4776 | 3.0 | 3297 | 0.3230 | 0.9125 |
| 0.4535 | 4.0 | 4396 | 0.2854 | 0.9205 |
| 0.4204 | 5.0 | 5495 | 0.2915 | 0.9169 |
| 0.3756 | 6.0 | 6594 | 0.2914 | 0.9193 |
| 0.3603 | 7.0 | 7693 | 0.2645 | 0.9272 |
| 0.2885 | 8.0 | 8792 | 0.2599 | 0.9280 |
| 0.2753 | 9.0 | 9891 | 0.2565 | 0.9292 |
| 0.2902 | 10.0 | 10990 | 0.2526 | 0.9292 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2