adam_VitB-p16-224-1e-4-batch_16_epoch_4_classes_24
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1310
- Accuracy: 0.9713
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.1921 | 0.07 | 100 | 0.1979 | 0.9440 |
0.0888 | 0.14 | 200 | 0.1824 | 0.9411 |
0.0672 | 0.21 | 300 | 0.1626 | 0.9440 |
0.1239 | 0.28 | 400 | 0.1495 | 0.9569 |
0.0779 | 0.35 | 500 | 0.1835 | 0.9497 |
0.0253 | 0.42 | 600 | 0.1516 | 0.9612 |
0.0154 | 0.49 | 700 | 0.1872 | 0.9526 |
0.0177 | 0.56 | 800 | 0.1847 | 0.9511 |
0.0633 | 0.63 | 900 | 0.1888 | 0.9468 |
0.0559 | 0.7 | 1000 | 0.1592 | 0.9641 |
0.0484 | 0.77 | 1100 | 0.1500 | 0.9569 |
0.0876 | 0.84 | 1200 | 0.1985 | 0.9440 |
0.0044 | 0.91 | 1300 | 0.0950 | 0.9698 |
0.0394 | 0.97 | 1400 | 0.1589 | 0.9612 |
0.0018 | 1.04 | 1500 | 0.1356 | 0.9641 |
0.0004 | 1.11 | 1600 | 0.1458 | 0.9655 |
0.025 | 1.18 | 1700 | 0.1248 | 0.9713 |
0.0117 | 1.25 | 1800 | 0.1419 | 0.9655 |
0.0348 | 1.32 | 1900 | 0.1110 | 0.9713 |
0.0021 | 1.39 | 2000 | 0.0957 | 0.9741 |
0.0006 | 1.46 | 2100 | 0.1621 | 0.9540 |
0.0018 | 1.53 | 2200 | 0.1056 | 0.9698 |
0.0008 | 1.6 | 2300 | 0.1713 | 0.9511 |
0.0359 | 1.67 | 2400 | 0.1412 | 0.9727 |
0.0003 | 1.74 | 2500 | 0.1753 | 0.9684 |
0.0003 | 1.81 | 2600 | 0.1128 | 0.9784 |
0.0004 | 1.88 | 2700 | 0.1268 | 0.9626 |
0.0322 | 1.95 | 2800 | 0.0970 | 0.9770 |
0.0344 | 2.02 | 2900 | 0.1139 | 0.9727 |
0.015 | 2.09 | 3000 | 0.1818 | 0.9612 |
0.0001 | 2.16 | 3100 | 0.0968 | 0.9770 |
0.0001 | 2.23 | 3200 | 0.1150 | 0.9756 |
0.0002 | 2.3 | 3300 | 0.1187 | 0.9756 |
0.0723 | 2.37 | 3400 | 0.1634 | 0.9641 |
0.0016 | 2.44 | 3500 | 0.1201 | 0.9698 |
0.0004 | 2.51 | 3600 | 0.1333 | 0.9713 |
0.03 | 2.58 | 3700 | 0.1412 | 0.9698 |
0.0005 | 2.65 | 3800 | 0.1149 | 0.9727 |
0.0002 | 2.72 | 3900 | 0.1599 | 0.9684 |
0.0059 | 2.79 | 4000 | 0.1110 | 0.9770 |
0.0001 | 2.86 | 4100 | 0.1090 | 0.9741 |
0.0001 | 2.92 | 4200 | 0.1094 | 0.9698 |
0.0001 | 2.99 | 4300 | 0.1148 | 0.9727 |
0.0001 | 3.06 | 4400 | 0.1231 | 0.9713 |
0.0001 | 3.13 | 4500 | 0.1173 | 0.9698 |
0.0002 | 3.2 | 4600 | 0.1268 | 0.9698 |
0.0001 | 3.27 | 4700 | 0.1207 | 0.9698 |
0.0001 | 3.34 | 4800 | 0.1208 | 0.9684 |
0.0001 | 3.41 | 4900 | 0.1203 | 0.9684 |
0.0001 | 3.48 | 5000 | 0.1215 | 0.9698 |
0.0001 | 3.55 | 5100 | 0.1217 | 0.9698 |
0.0001 | 3.62 | 5200 | 0.1227 | 0.9698 |
0.0001 | 3.69 | 5300 | 0.1226 | 0.9698 |
0.0001 | 3.76 | 5400 | 0.1226 | 0.9698 |
0.0001 | 3.83 | 5500 | 0.1218 | 0.9713 |
0.0001 | 3.9 | 5600 | 0.1309 | 0.9727 |
0.0001 | 3.97 | 5700 | 0.1310 | 0.9713 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 155