rmsprop_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: 0.1712
- Accuracy: 0.9684
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 |
---|---|---|---|---|
0.0876 | 0.07 | 100 | 0.1851 | 0.9483 |
0.117 | 0.14 | 200 | 0.2321 | 0.9339 |
0.0244 | 0.21 | 300 | 0.1376 | 0.9641 |
0.058 | 0.28 | 400 | 0.3501 | 0.9267 |
0.0159 | 0.35 | 500 | 0.2359 | 0.9425 |
0.0322 | 0.42 | 600 | 0.1792 | 0.9641 |
0.0245 | 0.49 | 700 | 0.2543 | 0.9483 |
0.0189 | 0.56 | 800 | 0.1764 | 0.9626 |
0.0528 | 0.63 | 900 | 0.2989 | 0.9497 |
0.0423 | 0.7 | 1000 | 0.2146 | 0.9583 |
0.0585 | 0.77 | 1100 | 0.2581 | 0.9425 |
0.002 | 0.84 | 1200 | 0.1778 | 0.9641 |
0.0131 | 0.91 | 1300 | 0.2760 | 0.9497 |
0.0889 | 0.97 | 1400 | 0.2059 | 0.9540 |
0.1212 | 1.04 | 1500 | 0.2592 | 0.9440 |
0.0003 | 1.11 | 1600 | 0.1900 | 0.9655 |
0.0884 | 1.18 | 1700 | 0.1622 | 0.9655 |
0.0188 | 1.25 | 1800 | 0.2284 | 0.9511 |
0.0002 | 1.32 | 1900 | 0.1840 | 0.9670 |
0.0108 | 1.39 | 2000 | 0.2478 | 0.9598 |
0.0003 | 1.46 | 2100 | 0.2207 | 0.9555 |
0.0183 | 1.53 | 2200 | 0.1800 | 0.9655 |
0.0119 | 1.6 | 2300 | 0.1976 | 0.9598 |
0.0407 | 1.67 | 2400 | 0.2089 | 0.9655 |
0.0001 | 1.74 | 2500 | 0.2273 | 0.9612 |
0.0005 | 1.81 | 2600 | 0.2895 | 0.9526 |
0.0048 | 1.88 | 2700 | 0.2115 | 0.9569 |
0.0391 | 1.95 | 2800 | 0.2026 | 0.9655 |
0.0001 | 2.02 | 2900 | 0.2276 | 0.9626 |
0.0108 | 2.09 | 3000 | 0.2089 | 0.9612 |
0.0 | 2.16 | 3100 | 0.2548 | 0.9583 |
0.0002 | 2.23 | 3200 | 0.2763 | 0.9626 |
0.0002 | 2.3 | 3300 | 0.1982 | 0.9655 |
0.0094 | 2.37 | 3400 | 0.2170 | 0.9655 |
0.0162 | 2.44 | 3500 | 0.1912 | 0.9655 |
0.0004 | 2.51 | 3600 | 0.2224 | 0.9655 |
0.0029 | 2.58 | 3700 | 0.1788 | 0.9713 |
0.0 | 2.65 | 3800 | 0.1954 | 0.9655 |
0.0107 | 2.72 | 3900 | 0.2269 | 0.9598 |
0.0001 | 2.79 | 4000 | 0.1996 | 0.9655 |
0.0001 | 2.86 | 4100 | 0.2232 | 0.9626 |
0.0 | 2.92 | 4200 | 0.1967 | 0.9713 |
0.0003 | 2.99 | 4300 | 0.1802 | 0.9655 |
0.0 | 3.06 | 4400 | 0.1779 | 0.9670 |
0.0 | 3.13 | 4500 | 0.1848 | 0.9655 |
0.0 | 3.2 | 4600 | 0.1849 | 0.9655 |
0.0 | 3.27 | 4700 | 0.1924 | 0.9641 |
0.0 | 3.34 | 4800 | 0.1802 | 0.9655 |
0.0 | 3.41 | 4900 | 0.1716 | 0.9698 |
0.0001 | 3.48 | 5000 | 0.1939 | 0.9670 |
0.0 | 3.55 | 5100 | 0.1850 | 0.9670 |
0.0 | 3.62 | 5200 | 0.1906 | 0.9684 |
0.0 | 3.69 | 5300 | 0.1909 | 0.9698 |
0.0 | 3.76 | 5400 | 0.1763 | 0.9698 |
0.0 | 3.83 | 5500 | 0.1718 | 0.9684 |
0.0 | 3.9 | 5600 | 0.1709 | 0.9684 |
0.0 | 3.97 | 5700 | 0.1712 | 0.9684 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 170