convnext-base-3e-5-batch-8
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: 0.2863
- Accuracy: 0.9379
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: 8
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|
| 0.5851 | 1.0 | 2198 | 0.3808 | 0.8918 |
| 0.3975 | 2.0 | 4396 | 0.3232 | 0.9093 |
| 0.3337 | 3.0 | 6594 | 0.3210 | 0.9252 |
| 0.2279 | 4.0 | 8792 | 0.3030 | 0.9308 |
| 0.1696 | 5.0 | 10990 | 0.3478 | 0.9292 |
| 0.1658 | 6.0 | 13188 | 0.3084 | 0.9427 |
| 0.1383 | 7.0 | 15386 | 0.3319 | 0.9392 |
| 0.1222 | 8.0 | 17584 | 0.3132 | 0.9479 |
| 0.1196 | 9.0 | 19782 | 0.3136 | 0.9467 |
| 0.1257 | 10.0 | 21980 | 0.3120 | 0.9483 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2