devout-voice-234
This model is a fine-tuned version of facebook/convnext-tiny-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4823
- Accuracy: 0.6195
- Precision: 0.7580
- Recall: 0.6195
- F1: 0.6418
- Roc Auc: 0.9108
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: 40
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Roc Auc |
|---|---|---|---|---|---|---|---|---|
| 1.3843 | 1.0 | 15 | 1.3708 | 0.2922 | 0.5011 | 0.2922 | 0.2957 | 0.6501 |
| 1.3354 | 2.0 | 30 | 1.3256 | 0.5062 | 0.5231 | 0.5062 | 0.4827 | 0.7450 |
| 1.2134 | 3.0 | 45 | 1.1394 | 0.5508 | 0.5749 | 0.5508 | 0.4751 | 0.7934 |
| 1.0408 | 4.0 | 60 | 0.9792 | 0.5188 | 0.6137 | 0.5188 | 0.5357 | 0.8100 |
| 0.881 | 5.0 | 75 | 0.6658 | 0.5508 | 0.5913 | 0.5508 | 0.5508 | 0.8320 |
| 0.7118 | 6.0 | 90 | 0.6165 | 0.5086 | 0.6760 | 0.5086 | 0.5172 | 0.8318 |
| 0.7556 | 7.0 | 105 | 0.5697 | 0.6070 | 0.6564 | 0.6070 | 0.6078 | 0.8671 |
| 0.6212 | 8.0 | 120 | 0.5433 | 0.5664 | 0.7000 | 0.5664 | 0.5755 | 0.8680 |
| 0.5591 | 9.0 | 135 | 0.4504 | 0.6797 | 0.7197 | 0.6797 | 0.6849 | 0.8983 |
| 0.4785 | 10.0 | 150 | 0.4269 | 0.6727 | 0.7115 | 0.6727 | 0.6706 | 0.9120 |
| 0.4093 | 11.0 | 165 | 0.4239 | 0.6742 | 0.7948 | 0.6742 | 0.6650 | 0.9345 |
| 0.4033 | 12.0 | 180 | 0.4823 | 0.6195 | 0.7580 | 0.6195 | 0.6418 | 0.9108 |
Framework versions
- Transformers 4.52.3
- Pytorch 2.7.0+cpu
- Datasets 3.6.0
- Tokenizers 0.21.0