smids_3x_deit_small_adamax_0001_fold3
This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.8015
- Accuracy: 0.905
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.3198 | 1.0 | 225 | 0.2665 | 0.8883 |
| 0.2096 | 2.0 | 450 | 0.2570 | 0.9017 |
| 0.0616 | 3.0 | 675 | 0.3183 | 0.9 |
| 0.0236 | 4.0 | 900 | 0.4110 | 0.8967 |
| 0.027 | 5.0 | 1125 | 0.4898 | 0.8833 |
| 0.0024 | 6.0 | 1350 | 0.5169 | 0.895 |
| 0.0276 | 7.0 | 1575 | 0.7378 | 0.89 |
| 0.0356 | 8.0 | 1800 | 0.5877 | 0.8983 |
| 0.0219 | 9.0 | 2025 | 0.6725 | 0.8967 |
| 0.0365 | 10.0 | 2250 | 0.7831 | 0.8833 |
| 0.0245 | 11.0 | 2475 | 0.6640 | 0.9033 |
| 0.0004 | 12.0 | 2700 | 0.7728 | 0.88 |
| 0.0002 | 13.0 | 2925 | 0.7409 | 0.8917 |
| 0.0001 | 14.0 | 3150 | 0.6940 | 0.9033 |
| 0.0 | 15.0 | 3375 | 0.7164 | 0.9 |
| 0.0 | 16.0 | 3600 | 0.7412 | 0.9033 |
| 0.0 | 17.0 | 3825 | 0.7630 | 0.9017 |
| 0.0001 | 18.0 | 4050 | 0.7681 | 0.9017 |
| 0.0 | 19.0 | 4275 | 0.7425 | 0.9033 |
| 0.0 | 20.0 | 4500 | 0.7631 | 0.8967 |
| 0.0 | 21.0 | 4725 | 0.7304 | 0.9067 |
| 0.0 | 22.0 | 4950 | 0.7313 | 0.9067 |
| 0.0 | 23.0 | 5175 | 0.7463 | 0.9 |
| 0.0 | 24.0 | 5400 | 0.7318 | 0.9083 |
| 0.0047 | 25.0 | 5625 | 0.7438 | 0.905 |
| 0.0 | 26.0 | 5850 | 0.7500 | 0.905 |
| 0.0 | 27.0 | 6075 | 0.7544 | 0.905 |
| 0.0 | 28.0 | 6300 | 0.7484 | 0.905 |
| 0.0046 | 29.0 | 6525 | 0.7585 | 0.9067 |
| 0.0039 | 30.0 | 6750 | 0.7608 | 0.905 |
| 0.0038 | 31.0 | 6975 | 0.7603 | 0.9033 |
| 0.0 | 32.0 | 7200 | 0.7834 | 0.9067 |
| 0.0 | 33.0 | 7425 | 0.7762 | 0.9033 |
| 0.0 | 34.0 | 7650 | 0.7871 | 0.9033 |
| 0.0 | 35.0 | 7875 | 0.7831 | 0.9067 |
| 0.0 | 36.0 | 8100 | 0.7821 | 0.9017 |
| 0.0 | 37.0 | 8325 | 0.7857 | 0.9067 |
| 0.0 | 38.0 | 8550 | 0.7857 | 0.9017 |
| 0.0 | 39.0 | 8775 | 0.7870 | 0.9033 |
| 0.0 | 40.0 | 9000 | 0.7883 | 0.9033 |
| 0.0 | 41.0 | 9225 | 0.7951 | 0.905 |
| 0.0 | 42.0 | 9450 | 0.7962 | 0.9033 |
| 0.0 | 43.0 | 9675 | 0.7972 | 0.9033 |
| 0.0 | 44.0 | 9900 | 0.7957 | 0.9017 |
| 0.0 | 45.0 | 10125 | 0.7978 | 0.9033 |
| 0.0 | 46.0 | 10350 | 0.8000 | 0.905 |
| 0.0025 | 47.0 | 10575 | 0.7996 | 0.905 |
| 0.0 | 48.0 | 10800 | 0.8016 | 0.905 |
| 0.0 | 49.0 | 11025 | 0.8022 | 0.905 |
| 0.0 | 50.0 | 11250 | 0.8015 | 0.905 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2