convnextv2-huge-1k-224-finetuned-cassava-leaf-disease
This model is a fine-tuned version of facebook/convnextv2-huge-1k-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3433
- Accuracy: 0.8897
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: 120
- eval_batch_size: 120
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 480
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 7.9894 | 0.25 | 10 | 4.9516 | 0.0206 |
| 2.7782 | 0.5 | 20 | 1.6759 | 0.6196 |
| 1.2699 | 0.75 | 30 | 0.9878 | 0.6626 |
| 0.8247 | 0.99 | 40 | 0.6755 | 0.7640 |
| 0.6353 | 1.24 | 50 | 0.5472 | 0.8079 |
| 0.5418 | 1.49 | 60 | 0.4924 | 0.8369 |
| 0.4577 | 1.74 | 70 | 0.4422 | 0.8537 |
| 0.4627 | 1.99 | 80 | 0.3943 | 0.8706 |
| 0.4235 | 2.24 | 90 | 0.3868 | 0.8715 |
| 0.4068 | 2.48 | 100 | 0.3879 | 0.8645 |
| 0.4088 | 2.73 | 110 | 0.4149 | 0.8579 |
| 0.3866 | 2.98 | 120 | 0.3489 | 0.8836 |
| 0.3776 | 3.23 | 130 | 0.3731 | 0.8743 |
| 0.3303 | 3.48 | 140 | 0.3719 | 0.8734 |
| 0.3548 | 3.73 | 150 | 0.3917 | 0.8668 |
| 0.3638 | 3.98 | 160 | 0.3561 | 0.8738 |
| 0.3292 | 4.22 | 170 | 0.3518 | 0.8855 |
| 0.3363 | 4.47 | 180 | 0.3561 | 0.8850 |
| 0.3123 | 4.72 | 190 | 0.3452 | 0.8794 |
| 0.3395 | 4.97 | 200 | 0.3385 | 0.8841 |
| 0.2851 | 5.22 | 210 | 0.3467 | 0.8883 |
| 0.3113 | 5.47 | 220 | 0.3393 | 0.8841 |
| 0.3035 | 5.71 | 230 | 0.3444 | 0.8785 |
| 0.3123 | 5.96 | 240 | 0.3321 | 0.8804 |
| 0.2683 | 6.21 | 250 | 0.3407 | 0.8813 |
| 0.2811 | 6.46 | 260 | 0.3396 | 0.8850 |
| 0.2779 | 6.71 | 270 | 0.3318 | 0.8869 |
| 0.2733 | 6.96 | 280 | 0.3342 | 0.8897 |
| 0.2661 | 7.2 | 290 | 0.3303 | 0.8916 |
| 0.2588 | 7.45 | 300 | 0.3387 | 0.8921 |
| 0.2586 | 7.7 | 310 | 0.3373 | 0.8888 |
| 0.2641 | 7.95 | 320 | 0.3328 | 0.8860 |
| 0.2408 | 8.2 | 330 | 0.3490 | 0.8818 |
| 0.2375 | 8.45 | 340 | 0.3419 | 0.8846 |
| 0.2507 | 8.7 | 350 | 0.3473 | 0.8874 |
| 0.2555 | 8.94 | 360 | 0.3382 | 0.8874 |
| 0.2299 | 9.19 | 370 | 0.3399 | 0.8888 |
| 0.2309 | 9.44 | 380 | 0.3415 | 0.8855 |
| 0.2344 | 9.69 | 390 | 0.3431 | 0.8897 |
| 0.2253 | 9.94 | 400 | 0.3433 | 0.8897 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.1
- Datasets 2.18.0
- Tokenizers 0.15.1