smids_5x_deit_tiny_sgd_001_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: 0.2974
- Accuracy: 0.8915
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 |
|---|---|---|---|---|
| 0.7214 | 1.0 | 376 | 0.7354 | 0.7112 |
| 0.5305 | 2.0 | 752 | 0.5484 | 0.7780 |
| 0.423 | 3.0 | 1128 | 0.4775 | 0.8063 |
| 0.4098 | 4.0 | 1504 | 0.4302 | 0.8364 |
| 0.4286 | 5.0 | 1880 | 0.4059 | 0.8497 |
| 0.3605 | 6.0 | 2256 | 0.3872 | 0.8548 |
| 0.3093 | 7.0 | 2632 | 0.3738 | 0.8648 |
| 0.3348 | 8.0 | 3008 | 0.3632 | 0.8664 |
| 0.3284 | 9.0 | 3384 | 0.3510 | 0.8765 |
| 0.3008 | 10.0 | 3760 | 0.3447 | 0.8748 |
| 0.289 | 11.0 | 4136 | 0.3398 | 0.8798 |
| 0.2542 | 12.0 | 4512 | 0.3320 | 0.8848 |
| 0.245 | 13.0 | 4888 | 0.3263 | 0.8865 |
| 0.2258 | 14.0 | 5264 | 0.3225 | 0.8865 |
| 0.3082 | 15.0 | 5640 | 0.3188 | 0.8848 |
| 0.2685 | 16.0 | 6016 | 0.3171 | 0.8848 |
| 0.2379 | 17.0 | 6392 | 0.3137 | 0.8865 |
| 0.2778 | 18.0 | 6768 | 0.3111 | 0.8848 |
| 0.2374 | 19.0 | 7144 | 0.3083 | 0.8848 |
| 0.1845 | 20.0 | 7520 | 0.3061 | 0.8848 |
| 0.2126 | 21.0 | 7896 | 0.3049 | 0.8865 |
| 0.2068 | 22.0 | 8272 | 0.3078 | 0.8831 |
| 0.2364 | 23.0 | 8648 | 0.3060 | 0.8798 |
| 0.1851 | 24.0 | 9024 | 0.3035 | 0.8881 |
| 0.2035 | 25.0 | 9400 | 0.3013 | 0.8848 |
| 0.2146 | 26.0 | 9776 | 0.3016 | 0.8881 |
| 0.1495 | 27.0 | 10152 | 0.2986 | 0.8915 |
| 0.1962 | 28.0 | 10528 | 0.2989 | 0.8898 |
| 0.2019 | 29.0 | 10904 | 0.2993 | 0.8881 |
| 0.1531 | 30.0 | 11280 | 0.2975 | 0.8932 |
| 0.1643 | 31.0 | 11656 | 0.2990 | 0.8898 |
| 0.2082 | 32.0 | 12032 | 0.2991 | 0.8881 |
| 0.1845 | 33.0 | 12408 | 0.2980 | 0.8915 |
| 0.1333 | 34.0 | 12784 | 0.2976 | 0.8932 |
| 0.1524 | 35.0 | 13160 | 0.3000 | 0.8865 |
| 0.1908 | 36.0 | 13536 | 0.2977 | 0.8915 |
| 0.1391 | 37.0 | 13912 | 0.2964 | 0.8948 |
| 0.1756 | 38.0 | 14288 | 0.2975 | 0.8915 |
| 0.2131 | 39.0 | 14664 | 0.2969 | 0.8932 |
| 0.1588 | 40.0 | 15040 | 0.2977 | 0.8898 |
| 0.1631 | 41.0 | 15416 | 0.2962 | 0.8932 |
| 0.1431 | 42.0 | 15792 | 0.2974 | 0.8915 |
| 0.1556 | 43.0 | 16168 | 0.2976 | 0.8898 |
| 0.1705 | 44.0 | 16544 | 0.2978 | 0.8915 |
| 0.1792 | 45.0 | 16920 | 0.2986 | 0.8898 |
| 0.1949 | 46.0 | 17296 | 0.2975 | 0.8915 |
| 0.1472 | 47.0 | 17672 | 0.2972 | 0.8915 |
| 0.139 | 48.0 | 18048 | 0.2974 | 0.8915 |
| 0.1452 | 49.0 | 18424 | 0.2974 | 0.8915 |
| 0.1388 | 50.0 | 18800 | 0.2974 | 0.8915 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2