hushem_5x_deit_small_sgd_00001_fold2
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: 1.4921
- Accuracy: 0.2
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 |
|---|---|---|---|---|
| 1.5247 | 1.0 | 27 | 1.5122 | 0.1778 |
| 1.5208 | 2.0 | 54 | 1.5113 | 0.1778 |
| 1.5431 | 3.0 | 81 | 1.5104 | 0.1778 |
| 1.5874 | 4.0 | 108 | 1.5095 | 0.1778 |
| 1.5185 | 5.0 | 135 | 1.5086 | 0.1778 |
| 1.5124 | 6.0 | 162 | 1.5078 | 0.1778 |
| 1.4656 | 7.0 | 189 | 1.5070 | 0.1778 |
| 1.5113 | 8.0 | 216 | 1.5062 | 0.1778 |
| 1.5043 | 9.0 | 243 | 1.5054 | 0.1778 |
| 1.505 | 10.0 | 270 | 1.5047 | 0.1778 |
| 1.4599 | 11.0 | 297 | 1.5040 | 0.1778 |
| 1.5036 | 12.0 | 324 | 1.5033 | 0.1778 |
| 1.5237 | 13.0 | 351 | 1.5026 | 0.1778 |
| 1.511 | 14.0 | 378 | 1.5019 | 0.1778 |
| 1.5324 | 15.0 | 405 | 1.5013 | 0.1778 |
| 1.5272 | 16.0 | 432 | 1.5007 | 0.1778 |
| 1.5263 | 17.0 | 459 | 1.5002 | 0.1778 |
| 1.4937 | 18.0 | 486 | 1.4996 | 0.1778 |
| 1.5117 | 19.0 | 513 | 1.4991 | 0.1778 |
| 1.516 | 20.0 | 540 | 1.4985 | 0.1778 |
| 1.5298 | 21.0 | 567 | 1.4981 | 0.1778 |
| 1.5031 | 22.0 | 594 | 1.4976 | 0.1778 |
| 1.496 | 23.0 | 621 | 1.4971 | 0.1778 |
| 1.4984 | 24.0 | 648 | 1.4967 | 0.2 |
| 1.4849 | 25.0 | 675 | 1.4963 | 0.2 |
| 1.5277 | 26.0 | 702 | 1.4959 | 0.2 |
| 1.4813 | 27.0 | 729 | 1.4955 | 0.2 |
| 1.5008 | 28.0 | 756 | 1.4952 | 0.2 |
| 1.5143 | 29.0 | 783 | 1.4948 | 0.2 |
| 1.5063 | 30.0 | 810 | 1.4945 | 0.2 |
| 1.5197 | 31.0 | 837 | 1.4942 | 0.2 |
| 1.4689 | 32.0 | 864 | 1.4940 | 0.2 |
| 1.5261 | 33.0 | 891 | 1.4937 | 0.2 |
| 1.5047 | 34.0 | 918 | 1.4935 | 0.2 |
| 1.4608 | 35.0 | 945 | 1.4933 | 0.2 |
| 1.5134 | 36.0 | 972 | 1.4931 | 0.2 |
| 1.4999 | 37.0 | 999 | 1.4929 | 0.2 |
| 1.4901 | 38.0 | 1026 | 1.4928 | 0.2 |
| 1.4933 | 39.0 | 1053 | 1.4926 | 0.2 |
| 1.5285 | 40.0 | 1080 | 1.4925 | 0.2 |
| 1.5189 | 41.0 | 1107 | 1.4924 | 0.2 |
| 1.5357 | 42.0 | 1134 | 1.4923 | 0.2 |
| 1.5726 | 43.0 | 1161 | 1.4923 | 0.2 |
| 1.4926 | 44.0 | 1188 | 1.4922 | 0.2 |
| 1.4915 | 45.0 | 1215 | 1.4922 | 0.2 |
| 1.4934 | 46.0 | 1242 | 1.4921 | 0.2 |
| 1.5214 | 47.0 | 1269 | 1.4921 | 0.2 |
| 1.5071 | 48.0 | 1296 | 1.4921 | 0.2 |
| 1.5711 | 49.0 | 1323 | 1.4921 | 0.2 |
| 1.4665 | 50.0 | 1350 | 1.4921 | 0.2 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0