convnextv2-tiny-1k-224-finetuned-barkley
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: 0.0083
- Precision: 1.0
- Recall: 1.0
- F1: 1.0
- Accuracy: 1.0
- Top1 Accuracy: 1.0
- Error Rate: 0.0
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | Top1 Accuracy | Error Rate |
|---|---|---|---|---|---|---|---|---|---|
| 1.4696 | 1.0 | 38 | 1.1807 | 0.7273 | 0.6513 | 0.6180 | 0.6768 | 0.6513 | 0.3232 |
| 0.7197 | 2.0 | 76 | 0.3719 | 0.9439 | 0.9408 | 0.9404 | 0.9434 | 0.9474 | 0.0566 |
| 0.2388 | 3.0 | 114 | 0.1489 | 0.9688 | 0.9671 | 0.9671 | 0.9716 | 0.9671 | 0.0284 |
| 0.1048 | 4.0 | 152 | 0.0730 | 0.9868 | 0.9868 | 0.9868 | 0.9878 | 0.9868 | 0.0122 |
| 0.1103 | 5.0 | 190 | 0.0288 | 0.9868 | 0.9868 | 0.9868 | 0.9878 | 0.9868 | 0.0122 |
| 0.072 | 6.0 | 228 | 0.0537 | 0.9877 | 0.9868 | 0.9869 | 0.9868 | 0.9868 | 0.0132 |
| 0.0248 | 7.0 | 266 | 0.0083 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 0.0 |
| 0.0371 | 8.0 | 304 | 0.0653 | 0.9819 | 0.9803 | 0.9802 | 0.9800 | 0.9803 | 0.0200 |
| 0.0626 | 9.0 | 342 | 0.2271 | 0.9545 | 0.9408 | 0.9404 | 0.95 | 0.9408 | 0.0500 |
| 0.07 | 10.0 | 380 | 0.0304 | 0.9936 | 0.9934 | 0.9934 | 0.9933 | 0.9934 | 0.0067 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.3.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1