<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
masked-pretraining-20-epoch
This model is a fine-tuned version of koobear/cam16-no-train-mask-final-50-epochs on the image_folder dataset. It achieves the following results on the evaluation set:
- Loss: 0.4113
- Accuracy: 0.9583
- F1: 0.9608
- Precision: 0.9666
- Recall: 0.9583
- Balanced Acc: 0.9594
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: 2e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | Balanced Acc |
---|---|---|---|---|---|---|---|---|
1.5138 | 1.0 | 6 | 1.4155 | 0.8125 | 0.8307 | 0.8854 | 0.8125 | 0.8785 |
1.3206 | 2.0 | 12 | 1.2731 | 0.8611 | 0.8688 | 0.8916 | 0.8611 | 0.8909 |
1.1665 | 3.0 | 18 | 1.1359 | 0.8958 | 0.9030 | 0.9214 | 0.8958 | 0.9052 |
1.0273 | 4.0 | 24 | 1.0128 | 0.9028 | 0.9090 | 0.9247 | 0.9028 | 0.9118 |
0.9014 | 5.0 | 30 | 0.9086 | 0.9097 | 0.9161 | 0.9342 | 0.9097 | 0.9217 |
0.7832 | 6.0 | 36 | 0.8094 | 0.9306 | 0.9356 | 0.9480 | 0.9306 | 0.9328 |
0.6802 | 7.0 | 42 | 0.7244 | 0.9375 | 0.9418 | 0.9513 | 0.9375 | 0.9394 |
0.5949 | 8.0 | 48 | 0.6657 | 0.9306 | 0.9348 | 0.9441 | 0.9306 | 0.9252 |
0.5295 | 9.0 | 54 | 0.6067 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.4758 | 10.0 | 60 | 0.5610 | 0.9444 | 0.9480 | 0.9556 | 0.9444 | 0.9461 |
0.4313 | 11.0 | 66 | 0.5307 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.3981 | 12.0 | 72 | 0.4942 | 0.9444 | 0.9480 | 0.9556 | 0.9444 | 0.9461 |
0.3711 | 13.0 | 78 | 0.4784 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.3436 | 14.0 | 84 | 0.4580 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.3232 | 15.0 | 90 | 0.4424 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.3096 | 16.0 | 96 | 0.4301 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.2996 | 17.0 | 102 | 0.4234 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.2898 | 18.0 | 108 | 0.4184 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.2845 | 19.0 | 114 | 0.4126 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
0.2826 | 20.0 | 120 | 0.4113 | 0.9583 | 0.9608 | 0.9666 | 0.9583 | 0.9594 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3