hushem_5x_deit_base_sgd_0001_fold4
This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.3872
- Accuracy: 0.1667
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: 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 |
|---|---|---|---|---|
| 1.4331 | 1.0 | 28 | 1.4140 | 0.1429 |
| 1.407 | 2.0 | 56 | 1.4126 | 0.1667 |
| 1.4098 | 3.0 | 84 | 1.4112 | 0.1667 |
| 1.4223 | 4.0 | 112 | 1.4100 | 0.1905 |
| 1.4411 | 5.0 | 140 | 1.4087 | 0.1905 |
| 1.4173 | 6.0 | 168 | 1.4075 | 0.1905 |
| 1.4276 | 7.0 | 196 | 1.4064 | 0.1905 |
| 1.4071 | 8.0 | 224 | 1.4054 | 0.1905 |
| 1.3997 | 9.0 | 252 | 1.4043 | 0.1905 |
| 1.3965 | 10.0 | 280 | 1.4034 | 0.1905 |
| 1.3989 | 11.0 | 308 | 1.4024 | 0.1667 |
| 1.3964 | 12.0 | 336 | 1.4014 | 0.1667 |
| 1.3857 | 13.0 | 364 | 1.4005 | 0.1667 |
| 1.3798 | 14.0 | 392 | 1.3997 | 0.1667 |
| 1.3951 | 15.0 | 420 | 1.3990 | 0.1667 |
| 1.4132 | 16.0 | 448 | 1.3982 | 0.1667 |
| 1.4034 | 17.0 | 476 | 1.3975 | 0.1667 |
| 1.3759 | 18.0 | 504 | 1.3968 | 0.1667 |
| 1.3925 | 19.0 | 532 | 1.3961 | 0.1667 |
| 1.3819 | 20.0 | 560 | 1.3955 | 0.1667 |
| 1.3696 | 21.0 | 588 | 1.3949 | 0.1667 |
| 1.369 | 22.0 | 616 | 1.3944 | 0.1667 |
| 1.3858 | 23.0 | 644 | 1.3938 | 0.1667 |
| 1.3737 | 24.0 | 672 | 1.3933 | 0.1905 |
| 1.3918 | 25.0 | 700 | 1.3928 | 0.1905 |
| 1.383 | 26.0 | 728 | 1.3923 | 0.1905 |
| 1.3591 | 27.0 | 756 | 1.3919 | 0.1905 |
| 1.373 | 28.0 | 784 | 1.3914 | 0.1905 |
| 1.3774 | 29.0 | 812 | 1.3910 | 0.1905 |
| 1.4107 | 30.0 | 840 | 1.3906 | 0.1905 |
| 1.3774 | 31.0 | 868 | 1.3903 | 0.1905 |
| 1.372 | 32.0 | 896 | 1.3899 | 0.1905 |
| 1.3576 | 33.0 | 924 | 1.3896 | 0.1905 |
| 1.3584 | 34.0 | 952 | 1.3892 | 0.1905 |
| 1.3749 | 35.0 | 980 | 1.3889 | 0.1905 |
| 1.3548 | 36.0 | 1008 | 1.3887 | 0.1905 |
| 1.3718 | 37.0 | 1036 | 1.3884 | 0.1905 |
| 1.3762 | 38.0 | 1064 | 1.3882 | 0.1667 |
| 1.367 | 39.0 | 1092 | 1.3880 | 0.1667 |
| 1.3764 | 40.0 | 1120 | 1.3878 | 0.1667 |
| 1.3651 | 41.0 | 1148 | 1.3877 | 0.1667 |
| 1.3606 | 42.0 | 1176 | 1.3876 | 0.1667 |
| 1.3658 | 43.0 | 1204 | 1.3874 | 0.1667 |
| 1.3704 | 44.0 | 1232 | 1.3874 | 0.1667 |
| 1.3557 | 45.0 | 1260 | 1.3873 | 0.1667 |
| 1.3731 | 46.0 | 1288 | 1.3872 | 0.1667 |
| 1.3663 | 47.0 | 1316 | 1.3872 | 0.1667 |
| 1.3426 | 48.0 | 1344 | 1.3872 | 0.1667 |
| 1.3656 | 49.0 | 1372 | 1.3872 | 0.1667 |
| 1.3695 | 50.0 | 1400 | 1.3872 | 0.1667 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0