convnextv2-base-1k-224-finetuned-cassava-leaf-disease
This model is a fine-tuned version of facebook/convnextv2-base-1k-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3847
- Accuracy: 0.8668
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: 300
- eval_batch_size: 300
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 1200
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 16
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4746 | 0.98 | 16 | 1.1574 | 0.6150 |
| 0.9672 | 1.97 | 32 | 0.8080 | 0.7084 |
| 0.7544 | 2.95 | 48 | 0.6316 | 0.7832 |
| 0.5568 | 4.0 | 65 | 0.5164 | 0.8350 |
| 0.4714 | 4.98 | 81 | 0.4622 | 0.8486 |
| 0.4361 | 5.97 | 97 | 0.4373 | 0.8509 |
| 0.412 | 6.95 | 113 | 0.4347 | 0.8570 |
| 0.3786 | 8.0 | 130 | 0.4091 | 0.8598 |
| 0.3741 | 8.98 | 146 | 0.4161 | 0.8598 |
| 0.3649 | 9.97 | 162 | 0.4062 | 0.8668 |
| 0.3532 | 10.95 | 178 | 0.3931 | 0.8668 |
| 0.3467 | 12.0 | 195 | 0.3884 | 0.8645 |
| 0.3379 | 12.98 | 211 | 0.3850 | 0.8645 |
| 0.3428 | 13.97 | 227 | 0.3911 | 0.8715 |
| 0.3409 | 14.95 | 243 | 0.3887 | 0.8687 |
| 0.3397 | 15.75 | 256 | 0.3847 | 0.8668 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1
- Datasets 2.18.0
- Tokenizers 0.15.1