smids_10x_beit_large_sgd_00001_fold5
This model is a fine-tuned version of microsoft/beit-large-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.8273
- Accuracy: 0.6317
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 |
|---|---|---|---|---|
| 1.2351 | 1.0 | 750 | 1.2384 | 0.335 |
| 1.188 | 2.0 | 1500 | 1.2067 | 0.3417 |
| 1.1425 | 3.0 | 2250 | 1.1794 | 0.35 |
| 1.0663 | 4.0 | 3000 | 1.1549 | 0.36 |
| 1.0302 | 5.0 | 3750 | 1.1332 | 0.3733 |
| 1.0803 | 6.0 | 4500 | 1.1133 | 0.3783 |
| 1.0194 | 7.0 | 5250 | 1.0948 | 0.395 |
| 1.041 | 8.0 | 6000 | 1.0776 | 0.4133 |
| 0.958 | 9.0 | 6750 | 1.0617 | 0.4267 |
| 0.9328 | 10.0 | 7500 | 1.0465 | 0.44 |
| 0.9293 | 11.0 | 8250 | 1.0324 | 0.4533 |
| 0.9087 | 12.0 | 9000 | 1.0189 | 0.465 |
| 0.9387 | 13.0 | 9750 | 1.0063 | 0.4783 |
| 0.8996 | 14.0 | 10500 | 0.9944 | 0.4933 |
| 0.8606 | 15.0 | 11250 | 0.9830 | 0.5083 |
| 0.8536 | 16.0 | 12000 | 0.9723 | 0.5117 |
| 0.8222 | 17.0 | 12750 | 0.9621 | 0.5217 |
| 0.8298 | 18.0 | 13500 | 0.9525 | 0.53 |
| 0.9106 | 19.0 | 14250 | 0.9434 | 0.54 |
| 0.8462 | 20.0 | 15000 | 0.9347 | 0.5483 |
| 0.8209 | 21.0 | 15750 | 0.9265 | 0.5533 |
| 0.8393 | 22.0 | 16500 | 0.9186 | 0.5583 |
| 0.8236 | 23.0 | 17250 | 0.9111 | 0.565 |
| 0.8476 | 24.0 | 18000 | 0.9042 | 0.5717 |
| 0.7925 | 25.0 | 18750 | 0.8975 | 0.5733 |
| 0.8294 | 26.0 | 19500 | 0.8913 | 0.5817 |
| 0.7415 | 27.0 | 20250 | 0.8856 | 0.585 |
| 0.8113 | 28.0 | 21000 | 0.8800 | 0.585 |
| 0.8087 | 29.0 | 21750 | 0.8747 | 0.5833 |
| 0.8087 | 30.0 | 22500 | 0.8698 | 0.59 |
| 0.7723 | 31.0 | 23250 | 0.8652 | 0.595 |
| 0.7864 | 32.0 | 24000 | 0.8609 | 0.6033 |
| 0.7882 | 33.0 | 24750 | 0.8569 | 0.6083 |
| 0.7814 | 34.0 | 25500 | 0.8532 | 0.61 |
| 0.8053 | 35.0 | 26250 | 0.8498 | 0.6117 |
| 0.7759 | 36.0 | 27000 | 0.8467 | 0.6167 |
| 0.73 | 37.0 | 27750 | 0.8438 | 0.6167 |
| 0.8437 | 38.0 | 28500 | 0.8412 | 0.6183 |
| 0.7621 | 39.0 | 29250 | 0.8389 | 0.6183 |
| 0.719 | 40.0 | 30000 | 0.8367 | 0.6217 |
| 0.7491 | 41.0 | 30750 | 0.8348 | 0.625 |
| 0.7887 | 42.0 | 31500 | 0.8332 | 0.625 |
| 0.8254 | 43.0 | 32250 | 0.8317 | 0.625 |
| 0.7337 | 44.0 | 33000 | 0.8305 | 0.6267 |
| 0.7762 | 45.0 | 33750 | 0.8295 | 0.6283 |
| 0.7277 | 46.0 | 34500 | 0.8286 | 0.6317 |
| 0.7733 | 47.0 | 35250 | 0.8280 | 0.6317 |
| 0.7249 | 48.0 | 36000 | 0.8276 | 0.6317 |
| 0.7591 | 49.0 | 36750 | 0.8274 | 0.6317 |
| 0.7103 | 50.0 | 37500 | 0.8273 | 0.6317 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2