smids_1x_deit_tiny_sgd_001_fold2
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: 0.4027
- Accuracy: 0.8303
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 |
|---|---|---|---|---|
| 1.092 | 1.0 | 75 | 1.0391 | 0.4642 |
| 0.9734 | 2.0 | 150 | 0.9266 | 0.5641 |
| 0.8567 | 3.0 | 225 | 0.8357 | 0.6339 |
| 0.7624 | 4.0 | 300 | 0.7647 | 0.6705 |
| 0.6877 | 5.0 | 375 | 0.7050 | 0.7205 |
| 0.6614 | 6.0 | 450 | 0.6522 | 0.7488 |
| 0.6494 | 7.0 | 525 | 0.6133 | 0.7604 |
| 0.6282 | 8.0 | 600 | 0.5876 | 0.7687 |
| 0.5404 | 9.0 | 675 | 0.5590 | 0.7787 |
| 0.5855 | 10.0 | 750 | 0.5391 | 0.7837 |
| 0.5304 | 11.0 | 825 | 0.5290 | 0.7820 |
| 0.4323 | 12.0 | 900 | 0.5141 | 0.7854 |
| 0.4483 | 13.0 | 975 | 0.4996 | 0.7987 |
| 0.3743 | 14.0 | 1050 | 0.4908 | 0.7970 |
| 0.3865 | 15.0 | 1125 | 0.4830 | 0.8003 |
| 0.4258 | 16.0 | 1200 | 0.4748 | 0.8053 |
| 0.3735 | 17.0 | 1275 | 0.4651 | 0.8087 |
| 0.4081 | 18.0 | 1350 | 0.4581 | 0.8087 |
| 0.4212 | 19.0 | 1425 | 0.4559 | 0.8120 |
| 0.3826 | 20.0 | 1500 | 0.4471 | 0.8170 |
| 0.4327 | 21.0 | 1575 | 0.4442 | 0.8170 |
| 0.3388 | 22.0 | 1650 | 0.4401 | 0.8136 |
| 0.3617 | 23.0 | 1725 | 0.4342 | 0.8186 |
| 0.3721 | 24.0 | 1800 | 0.4302 | 0.8220 |
| 0.3408 | 25.0 | 1875 | 0.4281 | 0.8170 |
| 0.2889 | 26.0 | 1950 | 0.4258 | 0.8203 |
| 0.362 | 27.0 | 2025 | 0.4230 | 0.8220 |
| 0.3321 | 28.0 | 2100 | 0.4203 | 0.8220 |
| 0.3436 | 29.0 | 2175 | 0.4171 | 0.8270 |
| 0.3045 | 30.0 | 2250 | 0.4153 | 0.8286 |
| 0.2874 | 31.0 | 2325 | 0.4166 | 0.8270 |
| 0.2803 | 32.0 | 2400 | 0.4140 | 0.8286 |
| 0.2889 | 33.0 | 2475 | 0.4102 | 0.8303 |
| 0.3058 | 34.0 | 2550 | 0.4088 | 0.8303 |
| 0.2713 | 35.0 | 2625 | 0.4095 | 0.8303 |
| 0.2862 | 36.0 | 2700 | 0.4078 | 0.8303 |
| 0.2924 | 37.0 | 2775 | 0.4065 | 0.8319 |
| 0.3044 | 38.0 | 2850 | 0.4070 | 0.8319 |
| 0.2574 | 39.0 | 2925 | 0.4050 | 0.8319 |
| 0.2876 | 40.0 | 3000 | 0.4054 | 0.8319 |
| 0.3084 | 41.0 | 3075 | 0.4046 | 0.8319 |
| 0.3077 | 42.0 | 3150 | 0.4046 | 0.8319 |
| 0.2848 | 43.0 | 3225 | 0.4035 | 0.8319 |
| 0.2711 | 44.0 | 3300 | 0.4038 | 0.8303 |
| 0.2493 | 45.0 | 3375 | 0.4031 | 0.8319 |
| 0.2526 | 46.0 | 3450 | 0.4030 | 0.8303 |
| 0.2813 | 47.0 | 3525 | 0.4028 | 0.8319 |
| 0.2908 | 48.0 | 3600 | 0.4027 | 0.8303 |
| 0.2758 | 49.0 | 3675 | 0.4027 | 0.8303 |
| 0.3051 | 50.0 | 3750 | 0.4027 | 0.8303 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0