convnext-base-wd1e-8-4e-5
This model is a fine-tuned version of facebook/convnextv2-base-22k-384 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2294
- Accuracy: 0.9468
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: 4e-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 |
|---|---|---|---|---|
| 0.6271 | 1.0 | 1099 | 0.3436 | 0.9054 |
| 0.4528 | 2.0 | 2198 | 0.2767 | 0.9256 |
| 0.3762 | 3.0 | 3297 | 0.2519 | 0.9268 |
| 0.2979 | 4.0 | 4396 | 0.2414 | 0.9372 |
| 0.2901 | 5.0 | 5495 | 0.2389 | 0.9427 |
| 0.2381 | 6.0 | 6594 | 0.2408 | 0.9419 |
| 0.2084 | 7.0 | 7693 | 0.2312 | 0.9463 |
| 0.1742 | 8.0 | 8792 | 0.2359 | 0.9451 |
| 0.1582 | 9.0 | 9891 | 0.2364 | 0.9479 |
| 0.1451 | 10.0 | 10990 | 0.2357 | 0.9495 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2