hushem_5x_deit_small_sgd_0001_fold1
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: 1.3695
- Accuracy: 0.3111
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 |
|---|---|---|---|---|
| 1.5195 | 1.0 | 27 | 1.4999 | 0.2889 |
| 1.4914 | 2.0 | 54 | 1.4892 | 0.2889 |
| 1.5108 | 3.0 | 81 | 1.4789 | 0.2889 |
| 1.5345 | 4.0 | 108 | 1.4698 | 0.2667 |
| 1.4684 | 5.0 | 135 | 1.4617 | 0.2667 |
| 1.4525 | 6.0 | 162 | 1.4534 | 0.2667 |
| 1.4298 | 7.0 | 189 | 1.4465 | 0.2667 |
| 1.4569 | 8.0 | 216 | 1.4397 | 0.2444 |
| 1.4283 | 9.0 | 243 | 1.4337 | 0.2444 |
| 1.4203 | 10.0 | 270 | 1.4280 | 0.2444 |
| 1.3871 | 11.0 | 297 | 1.4228 | 0.2444 |
| 1.4156 | 12.0 | 324 | 1.4180 | 0.2444 |
| 1.4346 | 13.0 | 351 | 1.4134 | 0.2444 |
| 1.4076 | 14.0 | 378 | 1.4093 | 0.2444 |
| 1.425 | 15.0 | 405 | 1.4059 | 0.2444 |
| 1.4406 | 16.0 | 432 | 1.4025 | 0.2444 |
| 1.4069 | 17.0 | 459 | 1.3996 | 0.2444 |
| 1.3779 | 18.0 | 486 | 1.3968 | 0.2444 |
| 1.3991 | 19.0 | 513 | 1.3941 | 0.2667 |
| 1.3962 | 20.0 | 540 | 1.3918 | 0.2667 |
| 1.3954 | 21.0 | 567 | 1.3897 | 0.2889 |
| 1.3886 | 22.0 | 594 | 1.3877 | 0.2889 |
| 1.3775 | 23.0 | 621 | 1.3858 | 0.2889 |
| 1.3714 | 24.0 | 648 | 1.3842 | 0.2889 |
| 1.4056 | 25.0 | 675 | 1.3826 | 0.2889 |
| 1.4026 | 26.0 | 702 | 1.3812 | 0.2889 |
| 1.359 | 27.0 | 729 | 1.3799 | 0.2889 |
| 1.3709 | 28.0 | 756 | 1.3787 | 0.2889 |
| 1.3667 | 29.0 | 783 | 1.3776 | 0.2889 |
| 1.3672 | 30.0 | 810 | 1.3766 | 0.2889 |
| 1.3762 | 31.0 | 837 | 1.3757 | 0.2889 |
| 1.3384 | 32.0 | 864 | 1.3749 | 0.2889 |
| 1.3698 | 33.0 | 891 | 1.3742 | 0.2889 |
| 1.3636 | 34.0 | 918 | 1.3735 | 0.3111 |
| 1.3439 | 35.0 | 945 | 1.3729 | 0.3111 |
| 1.3571 | 36.0 | 972 | 1.3723 | 0.3111 |
| 1.3688 | 37.0 | 999 | 1.3718 | 0.3111 |
| 1.3527 | 38.0 | 1026 | 1.3714 | 0.3111 |
| 1.3641 | 39.0 | 1053 | 1.3710 | 0.3111 |
| 1.3538 | 40.0 | 1080 | 1.3707 | 0.3111 |
| 1.3693 | 41.0 | 1107 | 1.3704 | 0.3111 |
| 1.3789 | 42.0 | 1134 | 1.3701 | 0.3111 |
| 1.3917 | 43.0 | 1161 | 1.3699 | 0.3111 |
| 1.3524 | 44.0 | 1188 | 1.3698 | 0.3111 |
| 1.367 | 45.0 | 1215 | 1.3696 | 0.3111 |
| 1.3553 | 46.0 | 1242 | 1.3696 | 0.3111 |
| 1.3523 | 47.0 | 1269 | 1.3695 | 0.3111 |
| 1.3646 | 48.0 | 1296 | 1.3695 | 0.3111 |
| 1.3891 | 49.0 | 1323 | 1.3695 | 0.3111 |
| 1.3396 | 50.0 | 1350 | 1.3695 | 0.3111 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0