hushem_40x_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: 2.3474
- Accuracy: 0.6889
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 |
|---|---|---|---|---|
| 0.0591 | 1.0 | 215 | 0.9478 | 0.7111 |
| 0.0016 | 2.0 | 430 | 1.0737 | 0.6889 |
| 0.0003 | 3.0 | 645 | 1.0732 | 0.7111 |
| 0.0001 | 4.0 | 860 | 1.2338 | 0.7111 |
| 0.0 | 5.0 | 1075 | 1.3886 | 0.7111 |
| 0.0 | 6.0 | 1290 | 1.5328 | 0.6889 |
| 0.0 | 7.0 | 1505 | 1.6761 | 0.6889 |
| 0.0 | 8.0 | 1720 | 1.8802 | 0.6889 |
| 0.0 | 9.0 | 1935 | 2.1375 | 0.6889 |
| 0.0 | 10.0 | 2150 | 2.2804 | 0.6889 |
| 0.0 | 11.0 | 2365 | 2.5018 | 0.6667 |
| 0.0 | 12.0 | 2580 | 2.6034 | 0.7111 |
| 0.0 | 13.0 | 2795 | 2.1119 | 0.7556 |
| 0.0 | 14.0 | 3010 | 2.5118 | 0.7111 |
| 0.0 | 15.0 | 3225 | 2.4215 | 0.6889 |
| 0.0 | 16.0 | 3440 | 2.4416 | 0.6889 |
| 0.0 | 17.0 | 3655 | 2.4789 | 0.6889 |
| 0.0 | 18.0 | 3870 | 2.5530 | 0.6889 |
| 0.0 | 19.0 | 4085 | 2.6223 | 0.6889 |
| 0.0 | 20.0 | 4300 | 2.7198 | 0.6889 |
| 0.0 | 21.0 | 4515 | 2.8171 | 0.7111 |
| 0.0 | 22.0 | 4730 | 2.8585 | 0.7111 |
| 0.0 | 23.0 | 4945 | 2.8584 | 0.7111 |
| 0.0 | 24.0 | 5160 | 2.7240 | 0.7111 |
| 0.0 | 25.0 | 5375 | 2.6522 | 0.7111 |
| 0.0 | 26.0 | 5590 | 2.6766 | 0.7111 |
| 0.0 | 27.0 | 5805 | 2.6051 | 0.7333 |
| 0.0 | 28.0 | 6020 | 2.4780 | 0.7333 |
| 0.0 | 29.0 | 6235 | 2.4371 | 0.7333 |
| 0.0 | 30.0 | 6450 | 2.3680 | 0.7333 |
| 0.0 | 31.0 | 6665 | 2.3696 | 0.7111 |
| 0.0 | 32.0 | 6880 | 2.3638 | 0.7333 |
| 0.0 | 33.0 | 7095 | 2.3261 | 0.7333 |
| 0.0 | 34.0 | 7310 | 2.3611 | 0.7333 |
| 0.0 | 35.0 | 7525 | 2.3737 | 0.7333 |
| 0.0 | 36.0 | 7740 | 2.3371 | 0.6889 |
| 0.0 | 37.0 | 7955 | 2.3450 | 0.7111 |
| 0.0 | 38.0 | 8170 | 2.3727 | 0.6889 |
| 0.0 | 39.0 | 8385 | 2.3620 | 0.6889 |
| 0.0 | 40.0 | 8600 | 2.3928 | 0.6889 |
| 0.0 | 41.0 | 8815 | 2.3547 | 0.6889 |
| 0.0 | 42.0 | 9030 | 2.3935 | 0.6889 |
| 0.0 | 43.0 | 9245 | 2.3835 | 0.6889 |
| 0.0 | 44.0 | 9460 | 2.3407 | 0.6889 |
| 0.0 | 45.0 | 9675 | 2.3628 | 0.6889 |
| 0.0 | 46.0 | 9890 | 2.3464 | 0.6889 |
| 0.0 | 47.0 | 10105 | 2.3571 | 0.6889 |
| 0.0 | 48.0 | 10320 | 2.3604 | 0.6889 |
| 0.0 | 49.0 | 10535 | 2.3495 | 0.6889 |
| 0.0 | 50.0 | 10750 | 2.3474 | 0.6889 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.1+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2