smids_1x_deit_tiny_rms_001_fold1
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: 1.2760
- Accuracy: 0.7479
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: 0.001
- 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 |
|---|---|---|---|---|
| 1.1876 | 1.0 | 76 | 1.3425 | 0.3356 |
| 1.1463 | 2.0 | 152 | 1.2111 | 0.3356 |
| 1.0954 | 3.0 | 228 | 1.0384 | 0.4908 |
| 0.9283 | 4.0 | 304 | 1.3266 | 0.3422 |
| 0.8828 | 5.0 | 380 | 0.8561 | 0.5609 |
| 0.9197 | 6.0 | 456 | 0.8658 | 0.5326 |
| 0.8712 | 7.0 | 532 | 0.8805 | 0.5242 |
| 0.8491 | 8.0 | 608 | 0.9865 | 0.5476 |
| 0.8029 | 9.0 | 684 | 0.8775 | 0.5609 |
| 0.7757 | 10.0 | 760 | 1.7456 | 0.4307 |
| 0.873 | 11.0 | 836 | 0.8686 | 0.5058 |
| 0.9608 | 12.0 | 912 | 0.7919 | 0.5893 |
| 0.7695 | 13.0 | 988 | 0.8046 | 0.5659 |
| 0.7753 | 14.0 | 1064 | 0.7823 | 0.5927 |
| 0.7166 | 15.0 | 1140 | 0.8383 | 0.5576 |
| 0.7186 | 16.0 | 1216 | 0.7798 | 0.6444 |
| 0.7445 | 17.0 | 1292 | 0.7722 | 0.6427 |
| 0.6602 | 18.0 | 1368 | 0.7838 | 0.6227 |
| 0.7252 | 19.0 | 1444 | 0.6881 | 0.6945 |
| 0.8232 | 20.0 | 1520 | 0.7249 | 0.6845 |
| 0.6331 | 21.0 | 1596 | 0.7098 | 0.6795 |
| 0.5883 | 22.0 | 1672 | 0.7178 | 0.6711 |
| 0.6215 | 23.0 | 1748 | 0.7089 | 0.6912 |
| 0.6142 | 24.0 | 1824 | 0.6619 | 0.7062 |
| 0.5494 | 25.0 | 1900 | 0.6904 | 0.7212 |
| 0.6135 | 26.0 | 1976 | 0.6508 | 0.7129 |
| 0.5009 | 27.0 | 2052 | 0.6608 | 0.7212 |
| 0.587 | 28.0 | 2128 | 0.7282 | 0.7112 |
| 0.4496 | 29.0 | 2204 | 0.6988 | 0.7095 |
| 0.4737 | 30.0 | 2280 | 0.7982 | 0.6728 |
| 0.5606 | 31.0 | 2356 | 0.7102 | 0.7062 |
| 0.5118 | 32.0 | 2432 | 0.6558 | 0.7479 |
| 0.4696 | 33.0 | 2508 | 0.9166 | 0.6978 |
| 0.5104 | 34.0 | 2584 | 0.7689 | 0.6995 |
| 0.4795 | 35.0 | 2660 | 0.6827 | 0.7362 |
| 0.432 | 36.0 | 2736 | 0.7633 | 0.7129 |
| 0.4887 | 37.0 | 2812 | 0.7070 | 0.7162 |
| 0.4102 | 38.0 | 2888 | 0.7227 | 0.7396 |
| 0.3613 | 39.0 | 2964 | 0.7184 | 0.7329 |
| 0.3288 | 40.0 | 3040 | 0.7694 | 0.7379 |
| 0.3339 | 41.0 | 3116 | 0.7145 | 0.7429 |
| 0.242 | 42.0 | 3192 | 0.8275 | 0.7429 |
| 0.3299 | 43.0 | 3268 | 0.8614 | 0.7446 |
| 0.2932 | 44.0 | 3344 | 0.9021 | 0.7362 |
| 0.2403 | 45.0 | 3420 | 0.9785 | 0.7479 |
| 0.1305 | 46.0 | 3496 | 1.0696 | 0.7546 |
| 0.1845 | 47.0 | 3572 | 1.1023 | 0.7529 |
| 0.1407 | 48.0 | 3648 | 1.2188 | 0.7513 |
| 0.0723 | 49.0 | 3724 | 1.2702 | 0.7479 |
| 0.0363 | 50.0 | 3800 | 1.2760 | 0.7479 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0