smids_1x_deit_small_rms_0001_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: 0.6765
- 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 |
|---|---|---|---|---|
| 1.1018 | 1.0 | 75 | 1.0371 | 0.4260 |
| 0.991 | 2.0 | 150 | 0.9921 | 0.4792 |
| 0.9572 | 3.0 | 225 | 0.9534 | 0.4692 |
| 0.9605 | 4.0 | 300 | 0.9410 | 0.4942 |
| 1.0177 | 5.0 | 375 | 0.9782 | 0.4792 |
| 0.8824 | 6.0 | 450 | 0.9530 | 0.4775 |
| 0.9937 | 7.0 | 525 | 1.2068 | 0.4143 |
| 0.9218 | 8.0 | 600 | 0.9562 | 0.4842 |
| 0.9543 | 9.0 | 675 | 0.9220 | 0.4892 |
| 0.9236 | 10.0 | 750 | 0.9222 | 0.4958 |
| 0.9252 | 11.0 | 825 | 0.8952 | 0.5075 |
| 0.8897 | 12.0 | 900 | 0.8977 | 0.5042 |
| 0.8737 | 13.0 | 975 | 0.8116 | 0.5691 |
| 0.8039 | 14.0 | 1050 | 0.7757 | 0.5790 |
| 0.7793 | 15.0 | 1125 | 0.8219 | 0.5824 |
| 0.8231 | 16.0 | 1200 | 0.7679 | 0.6057 |
| 0.8017 | 17.0 | 1275 | 0.7881 | 0.5874 |
| 0.7891 | 18.0 | 1350 | 0.8079 | 0.5990 |
| 0.7545 | 19.0 | 1425 | 0.7312 | 0.6456 |
| 0.7578 | 20.0 | 1500 | 0.7753 | 0.6123 |
| 0.8565 | 21.0 | 1575 | 0.7816 | 0.6073 |
| 0.7262 | 22.0 | 1650 | 0.8273 | 0.5840 |
| 0.7951 | 23.0 | 1725 | 0.7247 | 0.6339 |
| 0.7867 | 24.0 | 1800 | 0.7753 | 0.6173 |
| 0.7108 | 25.0 | 1875 | 0.7213 | 0.6805 |
| 0.6679 | 26.0 | 1950 | 0.7131 | 0.6556 |
| 0.7183 | 27.0 | 2025 | 0.7432 | 0.6456 |
| 0.6589 | 28.0 | 2100 | 0.6919 | 0.6938 |
| 0.6988 | 29.0 | 2175 | 0.7014 | 0.6689 |
| 0.6704 | 30.0 | 2250 | 0.6664 | 0.7038 |
| 0.6348 | 31.0 | 2325 | 0.6647 | 0.7038 |
| 0.6316 | 32.0 | 2400 | 0.6641 | 0.6988 |
| 0.5915 | 33.0 | 2475 | 0.6743 | 0.6839 |
| 0.6102 | 34.0 | 2550 | 0.6568 | 0.7038 |
| 0.5452 | 35.0 | 2625 | 0.6346 | 0.7271 |
| 0.5721 | 36.0 | 2700 | 0.6475 | 0.7255 |
| 0.5908 | 37.0 | 2775 | 0.6240 | 0.7388 |
| 0.6069 | 38.0 | 2850 | 0.6538 | 0.7354 |
| 0.4947 | 39.0 | 2925 | 0.6146 | 0.7438 |
| 0.4469 | 40.0 | 3000 | 0.6694 | 0.7038 |
| 0.5595 | 41.0 | 3075 | 0.5969 | 0.7438 |
| 0.524 | 42.0 | 3150 | 0.6251 | 0.7438 |
| 0.5223 | 43.0 | 3225 | 0.6144 | 0.7338 |
| 0.4414 | 44.0 | 3300 | 0.6374 | 0.7404 |
| 0.5093 | 45.0 | 3375 | 0.6328 | 0.7488 |
| 0.4116 | 46.0 | 3450 | 0.6556 | 0.7537 |
| 0.414 | 47.0 | 3525 | 0.6472 | 0.7604 |
| 0.445 | 48.0 | 3600 | 0.6566 | 0.7488 |
| 0.3661 | 49.0 | 3675 | 0.6775 | 0.7504 |
| 0.3935 | 50.0 | 3750 | 0.6765 | 0.7571 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0