hushem_1x_deit_tiny_rms_00001_fold2
This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.4676
- Accuracy: 0.6222
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: 1e-05
- 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 |
|---|---|---|---|---|
| No log | 1.0 | 6 | 1.3067 | 0.4222 |
| 1.3733 | 2.0 | 12 | 1.3951 | 0.4444 |
| 1.3733 | 3.0 | 18 | 1.3740 | 0.4222 |
| 0.6558 | 4.0 | 24 | 1.2467 | 0.5333 |
| 0.3343 | 5.0 | 30 | 1.5107 | 0.4667 |
| 0.3343 | 6.0 | 36 | 1.6079 | 0.4444 |
| 0.1446 | 7.0 | 42 | 1.2227 | 0.5333 |
| 0.1446 | 8.0 | 48 | 1.2018 | 0.5333 |
| 0.0575 | 9.0 | 54 | 1.2408 | 0.5111 |
| 0.0237 | 10.0 | 60 | 1.2581 | 0.5111 |
| 0.0237 | 11.0 | 66 | 1.4007 | 0.6 |
| 0.0072 | 12.0 | 72 | 1.2676 | 0.6444 |
| 0.0072 | 13.0 | 78 | 1.2933 | 0.5778 |
| 0.0036 | 14.0 | 84 | 1.3326 | 0.6222 |
| 0.0025 | 15.0 | 90 | 1.3074 | 0.6444 |
| 0.0025 | 16.0 | 96 | 1.3484 | 0.6222 |
| 0.002 | 17.0 | 102 | 1.3984 | 0.6222 |
| 0.002 | 18.0 | 108 | 1.3916 | 0.6222 |
| 0.0017 | 19.0 | 114 | 1.3871 | 0.6222 |
| 0.0014 | 20.0 | 120 | 1.4171 | 0.6222 |
| 0.0014 | 21.0 | 126 | 1.4207 | 0.6222 |
| 0.0012 | 22.0 | 132 | 1.4218 | 0.6222 |
| 0.0012 | 23.0 | 138 | 1.4371 | 0.6222 |
| 0.0011 | 24.0 | 144 | 1.4404 | 0.6222 |
| 0.001 | 25.0 | 150 | 1.4321 | 0.6222 |
| 0.001 | 26.0 | 156 | 1.4218 | 0.6222 |
| 0.0009 | 27.0 | 162 | 1.4367 | 0.6222 |
| 0.0009 | 28.0 | 168 | 1.4359 | 0.6222 |
| 0.0008 | 29.0 | 174 | 1.4387 | 0.6222 |
| 0.0008 | 30.0 | 180 | 1.4566 | 0.6222 |
| 0.0008 | 31.0 | 186 | 1.4528 | 0.6222 |
| 0.0007 | 32.0 | 192 | 1.4517 | 0.6222 |
| 0.0007 | 33.0 | 198 | 1.4535 | 0.6222 |
| 0.0007 | 34.0 | 204 | 1.4488 | 0.6444 |
| 0.0007 | 35.0 | 210 | 1.4494 | 0.6444 |
| 0.0007 | 36.0 | 216 | 1.4561 | 0.6444 |
| 0.0007 | 37.0 | 222 | 1.4595 | 0.6444 |
| 0.0007 | 38.0 | 228 | 1.4667 | 0.6222 |
| 0.0006 | 39.0 | 234 | 1.4671 | 0.6222 |
| 0.0007 | 40.0 | 240 | 1.4686 | 0.6222 |
| 0.0007 | 41.0 | 246 | 1.4681 | 0.6222 |
| 0.0006 | 42.0 | 252 | 1.4676 | 0.6222 |
| 0.0006 | 43.0 | 258 | 1.4676 | 0.6222 |
| 0.0006 | 44.0 | 264 | 1.4676 | 0.6222 |
| 0.0006 | 45.0 | 270 | 1.4676 | 0.6222 |
| 0.0006 | 46.0 | 276 | 1.4676 | 0.6222 |
| 0.0006 | 47.0 | 282 | 1.4676 | 0.6222 |
| 0.0006 | 48.0 | 288 | 1.4676 | 0.6222 |
| 0.0006 | 49.0 | 294 | 1.4676 | 0.6222 |
| 0.0006 | 50.0 | 300 | 1.4676 | 0.6222 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1