smids_5x_deit_small_rms_001_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: 0.5587
- Accuracy: 0.7571
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 |
|---|---|---|---|---|
| 1.1114 | 1.0 | 375 | 1.0890 | 0.3910 |
| 0.8594 | 2.0 | 750 | 0.8338 | 0.5258 |
| 0.7942 | 3.0 | 1125 | 0.8187 | 0.5857 |
| 0.8295 | 4.0 | 1500 | 0.8518 | 0.5441 |
| 0.7679 | 5.0 | 1875 | 0.8037 | 0.5574 |
| 0.78 | 6.0 | 2250 | 0.8086 | 0.5641 |
| 0.8051 | 7.0 | 2625 | 0.8475 | 0.5707 |
| 0.7931 | 8.0 | 3000 | 0.7986 | 0.5740 |
| 0.787 | 9.0 | 3375 | 0.8111 | 0.6273 |
| 0.7865 | 10.0 | 3750 | 0.7854 | 0.5874 |
| 0.7545 | 11.0 | 4125 | 0.7499 | 0.6156 |
| 0.8266 | 12.0 | 4500 | 0.7256 | 0.6373 |
| 0.699 | 13.0 | 4875 | 0.7372 | 0.6456 |
| 0.7725 | 14.0 | 5250 | 0.7195 | 0.6439 |
| 0.7834 | 15.0 | 5625 | 0.7235 | 0.6489 |
| 0.6954 | 16.0 | 6000 | 0.7061 | 0.6589 |
| 0.6911 | 17.0 | 6375 | 0.8229 | 0.5973 |
| 0.667 | 18.0 | 6750 | 0.6746 | 0.6772 |
| 0.6889 | 19.0 | 7125 | 0.6831 | 0.6955 |
| 0.6628 | 20.0 | 7500 | 0.6921 | 0.6922 |
| 0.7228 | 21.0 | 7875 | 0.6764 | 0.6656 |
| 0.7022 | 22.0 | 8250 | 0.6797 | 0.6905 |
| 0.6549 | 23.0 | 8625 | 0.6709 | 0.6772 |
| 0.7183 | 24.0 | 9000 | 0.6429 | 0.6955 |
| 0.6612 | 25.0 | 9375 | 0.6503 | 0.6988 |
| 0.6901 | 26.0 | 9750 | 0.7018 | 0.6739 |
| 0.7038 | 27.0 | 10125 | 0.6168 | 0.7271 |
| 0.6364 | 28.0 | 10500 | 0.6219 | 0.7121 |
| 0.6477 | 29.0 | 10875 | 0.6546 | 0.7188 |
| 0.5753 | 30.0 | 11250 | 0.6252 | 0.7221 |
| 0.6932 | 31.0 | 11625 | 0.6174 | 0.7271 |
| 0.6245 | 32.0 | 12000 | 0.6281 | 0.7255 |
| 0.6083 | 33.0 | 12375 | 0.6211 | 0.7105 |
| 0.6277 | 34.0 | 12750 | 0.5911 | 0.7471 |
| 0.596 | 35.0 | 13125 | 0.5943 | 0.7304 |
| 0.5539 | 36.0 | 13500 | 0.5798 | 0.7121 |
| 0.6231 | 37.0 | 13875 | 0.5842 | 0.7321 |
| 0.5692 | 38.0 | 14250 | 0.5897 | 0.7288 |
| 0.5587 | 39.0 | 14625 | 0.6220 | 0.7155 |
| 0.5891 | 40.0 | 15000 | 0.6063 | 0.7338 |
| 0.561 | 41.0 | 15375 | 0.5930 | 0.7338 |
| 0.5901 | 42.0 | 15750 | 0.5990 | 0.7288 |
| 0.5194 | 43.0 | 16125 | 0.5632 | 0.7488 |
| 0.5311 | 44.0 | 16500 | 0.5715 | 0.7488 |
| 0.5414 | 45.0 | 16875 | 0.5640 | 0.7537 |
| 0.5291 | 46.0 | 17250 | 0.5674 | 0.7504 |
| 0.5724 | 47.0 | 17625 | 0.5765 | 0.7454 |
| 0.4849 | 48.0 | 18000 | 0.5625 | 0.7438 |
| 0.5463 | 49.0 | 18375 | 0.5558 | 0.7571 |
| 0.4844 | 50.0 | 18750 | 0.5587 | 0.7571 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2