smids_1x_deit_tiny_sgd_00001_fold4
This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.1991
- Accuracy: 0.3667
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: 1e-05
- 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 |
|---|---|---|---|---|
| 1.3246 | 1.0 | 75 | 1.3568 | 0.3467 |
| 1.3026 | 2.0 | 150 | 1.3478 | 0.3467 |
| 1.3075 | 3.0 | 225 | 1.3392 | 0.3467 |
| 1.3769 | 4.0 | 300 | 1.3310 | 0.3467 |
| 1.2997 | 5.0 | 375 | 1.3231 | 0.3467 |
| 1.2661 | 6.0 | 450 | 1.3158 | 0.3433 |
| 1.2888 | 7.0 | 525 | 1.3087 | 0.3433 |
| 1.2851 | 8.0 | 600 | 1.3020 | 0.3433 |
| 1.2991 | 9.0 | 675 | 1.2955 | 0.3433 |
| 1.3514 | 10.0 | 750 | 1.2893 | 0.345 |
| 1.256 | 11.0 | 825 | 1.2833 | 0.3417 |
| 1.2501 | 12.0 | 900 | 1.2778 | 0.3417 |
| 1.2581 | 13.0 | 975 | 1.2726 | 0.345 |
| 1.279 | 14.0 | 1050 | 1.2675 | 0.345 |
| 1.281 | 15.0 | 1125 | 1.2628 | 0.345 |
| 1.2242 | 16.0 | 1200 | 1.2582 | 0.3483 |
| 1.1785 | 17.0 | 1275 | 1.2539 | 0.3483 |
| 1.2882 | 18.0 | 1350 | 1.2497 | 0.35 |
| 1.2177 | 19.0 | 1425 | 1.2459 | 0.3533 |
| 1.1848 | 20.0 | 1500 | 1.2422 | 0.3567 |
| 1.2931 | 21.0 | 1575 | 1.2388 | 0.3583 |
| 1.2179 | 22.0 | 1650 | 1.2355 | 0.3567 |
| 1.2465 | 23.0 | 1725 | 1.2324 | 0.3567 |
| 1.2403 | 24.0 | 1800 | 1.2294 | 0.355 |
| 1.2116 | 25.0 | 1875 | 1.2267 | 0.355 |
| 1.2221 | 26.0 | 1950 | 1.2242 | 0.36 |
| 1.167 | 27.0 | 2025 | 1.2218 | 0.36 |
| 1.2147 | 28.0 | 2100 | 1.2195 | 0.3583 |
| 1.2367 | 29.0 | 2175 | 1.2174 | 0.355 |
| 1.2142 | 30.0 | 2250 | 1.2154 | 0.355 |
| 1.2312 | 31.0 | 2325 | 1.2136 | 0.3533 |
| 1.1773 | 32.0 | 2400 | 1.2119 | 0.3517 |
| 1.1658 | 33.0 | 2475 | 1.2103 | 0.3517 |
| 1.2038 | 34.0 | 2550 | 1.2088 | 0.355 |
| 1.1521 | 35.0 | 2625 | 1.2075 | 0.3567 |
| 1.1878 | 36.0 | 2700 | 1.2062 | 0.3633 |
| 1.2013 | 37.0 | 2775 | 1.2051 | 0.365 |
| 1.1943 | 38.0 | 2850 | 1.2041 | 0.3633 |
| 1.1839 | 39.0 | 2925 | 1.2032 | 0.3667 |
| 1.1836 | 40.0 | 3000 | 1.2024 | 0.3683 |
| 1.1971 | 41.0 | 3075 | 1.2017 | 0.3683 |
| 1.1901 | 42.0 | 3150 | 1.2011 | 0.365 |
| 1.2156 | 43.0 | 3225 | 1.2005 | 0.365 |
| 1.2062 | 44.0 | 3300 | 1.2001 | 0.365 |
| 1.1956 | 45.0 | 3375 | 1.1998 | 0.365 |
| 1.2469 | 46.0 | 3450 | 1.1995 | 0.3633 |
| 1.1737 | 47.0 | 3525 | 1.1993 | 0.3667 |
| 1.1496 | 48.0 | 3600 | 1.1992 | 0.3667 |
| 1.1899 | 49.0 | 3675 | 1.1991 | 0.3667 |
| 1.2185 | 50.0 | 3750 | 1.1991 | 0.3667 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0