ConvNeXtV2_Tiny_v4
This model is a fine-tuned version of facebook/convnextv2-tiny-1k-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0577
- Accuracy: 0.9853
- Precision: 0.9871
- Recall: 0.9811
- F1: 0.9841
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: 0.0001
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 66
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.1899 | 1.0 | 111 | 0.2023 | 0.8999 | 0.8315 | 0.9823 | 0.9006 |
| 0.1830 | 2.0 | 222 | 0.0854 | 0.9814 | 0.9894 | 0.9701 | 0.9797 |
| 0.1961 | 3.0 | 333 | 0.0992 | 0.9721 | 0.9589 | 0.9817 | 0.9701 |
| 0.1580 | 4.0 | 444 | 0.0681 | 0.9839 | 0.9877 | 0.9774 | 0.9825 |
| 0.1596 | 5.0 | 555 | 0.0650 | 0.9848 | 0.9889 | 0.9780 | 0.9834 |
| 0.1220 | 6.0 | 666 | 0.0577 | 0.9853 | 0.9871 | 0.9811 | 0.9841 |
Framework versions
- Transformers 5.2.0
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.2