smids_1x_deit_small_sgd_0001_fold1
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: 0.9238
- Accuracy: 0.6194
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 |
|---|---|---|---|---|
| 1.1708 | 1.0 | 76 | 1.1671 | 0.2154 |
| 1.1329 | 2.0 | 152 | 1.1541 | 0.2421 |
| 1.1491 | 3.0 | 228 | 1.1427 | 0.2538 |
| 1.1383 | 4.0 | 304 | 1.1321 | 0.2654 |
| 1.095 | 5.0 | 380 | 1.1227 | 0.2838 |
| 1.1047 | 6.0 | 456 | 1.1138 | 0.3055 |
| 1.1073 | 7.0 | 532 | 1.1055 | 0.3322 |
| 1.1015 | 8.0 | 608 | 1.0974 | 0.3456 |
| 1.0955 | 9.0 | 684 | 1.0899 | 0.3639 |
| 1.0349 | 10.0 | 760 | 1.0825 | 0.3706 |
| 1.0617 | 11.0 | 836 | 1.0752 | 0.3957 |
| 1.0611 | 12.0 | 912 | 1.0681 | 0.4140 |
| 1.0517 | 13.0 | 988 | 1.0610 | 0.4240 |
| 1.0458 | 14.0 | 1064 | 1.0541 | 0.4357 |
| 1.0495 | 15.0 | 1140 | 1.0471 | 0.4391 |
| 1.032 | 16.0 | 1216 | 1.0402 | 0.4457 |
| 1.0199 | 17.0 | 1292 | 1.0334 | 0.4608 |
| 1.0216 | 18.0 | 1368 | 1.0267 | 0.4691 |
| 1.0137 | 19.0 | 1444 | 1.0202 | 0.4825 |
| 1.0117 | 20.0 | 1520 | 1.0136 | 0.5058 |
| 0.9951 | 21.0 | 1596 | 1.0073 | 0.5142 |
| 0.9978 | 22.0 | 1672 | 1.0011 | 0.5175 |
| 0.9715 | 23.0 | 1748 | 0.9952 | 0.5242 |
| 0.9775 | 24.0 | 1824 | 0.9895 | 0.5392 |
| 0.9696 | 25.0 | 1900 | 0.9841 | 0.5409 |
| 0.9601 | 26.0 | 1976 | 0.9789 | 0.5492 |
| 0.9807 | 27.0 | 2052 | 0.9740 | 0.5593 |
| 0.9357 | 28.0 | 2128 | 0.9694 | 0.5626 |
| 0.9396 | 29.0 | 2204 | 0.9650 | 0.5710 |
| 0.9629 | 30.0 | 2280 | 0.9608 | 0.5743 |
| 0.9473 | 31.0 | 2356 | 0.9570 | 0.5826 |
| 0.9153 | 32.0 | 2432 | 0.9532 | 0.5860 |
| 0.9343 | 33.0 | 2508 | 0.9497 | 0.5927 |
| 0.953 | 34.0 | 2584 | 0.9465 | 0.5993 |
| 0.949 | 35.0 | 2660 | 0.9435 | 0.6027 |
| 0.9108 | 36.0 | 2736 | 0.9407 | 0.6043 |
| 0.9432 | 37.0 | 2812 | 0.9382 | 0.6060 |
| 0.9019 | 38.0 | 2888 | 0.9358 | 0.6093 |
| 0.9269 | 39.0 | 2964 | 0.9337 | 0.6093 |
| 0.9369 | 40.0 | 3040 | 0.9318 | 0.6110 |
| 0.8967 | 41.0 | 3116 | 0.9301 | 0.6110 |
| 0.9379 | 42.0 | 3192 | 0.9286 | 0.6144 |
| 0.8809 | 43.0 | 3268 | 0.9273 | 0.6160 |
| 0.9188 | 44.0 | 3344 | 0.9263 | 0.6177 |
| 0.8925 | 45.0 | 3420 | 0.9254 | 0.6177 |
| 0.8985 | 46.0 | 3496 | 0.9247 | 0.6177 |
| 0.9127 | 47.0 | 3572 | 0.9242 | 0.6194 |
| 0.9099 | 48.0 | 3648 | 0.9239 | 0.6194 |
| 0.9072 | 49.0 | 3724 | 0.9238 | 0.6194 |
| 0.9001 | 50.0 | 3800 | 0.9238 | 0.6194 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0