bumbling-universe-237
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: 0.0992
- Accuracy: 0.9689
- Precision: 0.9689
- Recall: 0.9689
- F1: 0.9689
- Roc Auc: 0.9993
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: 256
- eval_batch_size: 256
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Roc Auc |
|---|---|---|---|---|---|---|---|---|
| 1.2837 | 1.0 | 161 | 0.2468 | 0.9334 | 0.9340 | 0.9334 | 0.9332 | 0.9972 |
| 0.5753 | 2.0 | 322 | 0.1520 | 0.9526 | 0.9533 | 0.9526 | 0.9525 | 0.9988 |
| 0.4552 | 3.0 | 483 | 0.1131 | 0.9646 | 0.9647 | 0.9646 | 0.9646 | 0.9991 |
| 0.3917 | 4.0 | 644 | 0.1026 | 0.9659 | 0.9660 | 0.9659 | 0.9659 | 0.9993 |
| 0.361 | 5.0 | 805 | 0.0992 | 0.9689 | 0.9689 | 0.9689 | 0.9689 | 0.9993 |
Framework versions
- Transformers 4.52.3
- Pytorch 2.7.0+cpu
- Datasets 3.6.0
- Tokenizers 0.21.0