10-convnext-base-224-finetuned-spiderTraining20-500
This model is a fine-tuned version of facebook/convnext-base-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2830
- Accuracy: 0.9319
- Precision: 0.9314
- Recall: 0.9294
- F1: 0.9297
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: 0.0005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- 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: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.7804 | 1.0 | 125 | 0.6029 | 0.8138 | 0.8234 | 0.8062 | 0.8065 |
| 0.5493 | 2.0 | 250 | 0.4794 | 0.8509 | 0.8576 | 0.8441 | 0.8432 |
| 0.4377 | 3.0 | 375 | 0.3774 | 0.8839 | 0.8861 | 0.8801 | 0.8799 |
| 0.2574 | 4.0 | 500 | 0.3446 | 0.9109 | 0.9106 | 0.9083 | 0.9078 |
| 0.3427 | 5.0 | 625 | 0.3314 | 0.8959 | 0.8941 | 0.8942 | 0.8918 |
| 0.2143 | 6.0 | 750 | 0.3146 | 0.9209 | 0.9257 | 0.9167 | 0.9191 |
| 0.1787 | 7.0 | 875 | 0.2863 | 0.9279 | 0.9261 | 0.9240 | 0.9240 |
| 0.1589 | 8.0 | 1000 | 0.2817 | 0.9309 | 0.9310 | 0.9296 | 0.9294 |
| 0.1332 | 9.0 | 1125 | 0.2893 | 0.9299 | 0.9298 | 0.9265 | 0.9271 |
| 0.1405 | 10.0 | 1250 | 0.2830 | 0.9319 | 0.9314 | 0.9294 | 0.9297 |
Framework versions
- Transformers 4.55.2
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.21.4