hushem_5x_deit_small_rms_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: 4.9688
- Accuracy: 0.4444
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 |
|---|---|---|---|---|
| 1.7253 | 1.0 | 27 | 1.4539 | 0.2444 |
| 1.4329 | 2.0 | 54 | 1.3827 | 0.3556 |
| 1.413 | 3.0 | 81 | 1.4287 | 0.2667 |
| 1.2575 | 4.0 | 108 | 1.2729 | 0.4222 |
| 1.2908 | 5.0 | 135 | 1.8913 | 0.3333 |
| 1.1882 | 6.0 | 162 | 1.1330 | 0.5111 |
| 1.0961 | 7.0 | 189 | 1.6635 | 0.3778 |
| 1.0705 | 8.0 | 216 | 1.0816 | 0.5556 |
| 0.8596 | 9.0 | 243 | 2.1258 | 0.4 |
| 0.8047 | 10.0 | 270 | 1.2784 | 0.4444 |
| 0.7923 | 11.0 | 297 | 2.1314 | 0.3778 |
| 0.7354 | 12.0 | 324 | 1.5632 | 0.3778 |
| 0.7076 | 13.0 | 351 | 1.6923 | 0.4 |
| 0.7272 | 14.0 | 378 | 1.4002 | 0.4222 |
| 0.6359 | 15.0 | 405 | 1.6646 | 0.4 |
| 0.5977 | 16.0 | 432 | 1.6603 | 0.4444 |
| 0.6463 | 17.0 | 459 | 1.5891 | 0.4444 |
| 0.6624 | 18.0 | 486 | 1.8543 | 0.4 |
| 0.5726 | 19.0 | 513 | 1.5545 | 0.5111 |
| 0.5713 | 20.0 | 540 | 1.7099 | 0.4 |
| 0.5626 | 21.0 | 567 | 1.6364 | 0.4 |
| 0.5358 | 22.0 | 594 | 1.8888 | 0.4667 |
| 0.5334 | 23.0 | 621 | 1.9170 | 0.4667 |
| 0.4645 | 24.0 | 648 | 2.0287 | 0.4222 |
| 0.5514 | 25.0 | 675 | 1.5224 | 0.4889 |
| 0.5254 | 26.0 | 702 | 2.4633 | 0.3556 |
| 0.441 | 27.0 | 729 | 1.7933 | 0.4222 |
| 0.3855 | 28.0 | 756 | 2.4673 | 0.4222 |
| 0.4099 | 29.0 | 783 | 2.6353 | 0.4222 |
| 0.4294 | 30.0 | 810 | 2.2588 | 0.4444 |
| 0.3329 | 31.0 | 837 | 2.3858 | 0.4 |
| 0.3787 | 32.0 | 864 | 2.2861 | 0.3333 |
| 0.3457 | 33.0 | 891 | 2.1705 | 0.4222 |
| 0.2509 | 34.0 | 918 | 2.4731 | 0.4222 |
| 0.1898 | 35.0 | 945 | 2.9376 | 0.3111 |
| 0.197 | 36.0 | 972 | 3.2201 | 0.4222 |
| 0.1255 | 37.0 | 999 | 2.5816 | 0.5333 |
| 0.1258 | 38.0 | 1026 | 3.6398 | 0.4 |
| 0.1837 | 39.0 | 1053 | 3.2179 | 0.4222 |
| 0.0881 | 40.0 | 1080 | 3.5990 | 0.4444 |
| 0.0547 | 41.0 | 1107 | 4.2943 | 0.4222 |
| 0.0315 | 42.0 | 1134 | 4.0730 | 0.4222 |
| 0.0187 | 43.0 | 1161 | 4.2944 | 0.4667 |
| 0.0043 | 44.0 | 1188 | 4.5081 | 0.4667 |
| 0.0016 | 45.0 | 1215 | 4.8996 | 0.4444 |
| 0.0009 | 46.0 | 1242 | 4.8993 | 0.4444 |
| 0.0009 | 47.0 | 1269 | 4.9469 | 0.4444 |
| 0.0007 | 48.0 | 1296 | 4.9681 | 0.4444 |
| 0.0006 | 49.0 | 1323 | 4.9688 | 0.4444 |
| 0.0006 | 50.0 | 1350 | 4.9688 | 0.4444 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0