hushem_40x_deit_base_sgd_00001_fold1
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: 1.4455
- Accuracy: 0.2889
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 |
---|---|---|---|---|
1.379 | 1.0 | 215 | 1.4681 | 0.2889 |
1.3967 | 2.0 | 430 | 1.4670 | 0.2889 |
1.423 | 3.0 | 645 | 1.4660 | 0.2889 |
1.4018 | 4.0 | 860 | 1.4650 | 0.2889 |
1.3899 | 5.0 | 1075 | 1.4640 | 0.2889 |
1.4076 | 6.0 | 1290 | 1.4631 | 0.2889 |
1.3743 | 7.0 | 1505 | 1.4622 | 0.2889 |
1.3724 | 8.0 | 1720 | 1.4613 | 0.2889 |
1.3757 | 9.0 | 1935 | 1.4604 | 0.2889 |
1.3783 | 10.0 | 2150 | 1.4596 | 0.2889 |
1.4141 | 11.0 | 2365 | 1.4589 | 0.2889 |
1.3702 | 12.0 | 2580 | 1.4581 | 0.2889 |
1.3842 | 13.0 | 2795 | 1.4574 | 0.2889 |
1.3926 | 14.0 | 3010 | 1.4567 | 0.2889 |
1.3764 | 15.0 | 3225 | 1.4560 | 0.2889 |
1.3955 | 16.0 | 3440 | 1.4553 | 0.2889 |
1.3752 | 17.0 | 3655 | 1.4547 | 0.2889 |
1.3872 | 18.0 | 3870 | 1.4541 | 0.2889 |
1.3795 | 19.0 | 4085 | 1.4535 | 0.2889 |
1.3768 | 20.0 | 4300 | 1.4530 | 0.2889 |
1.3609 | 21.0 | 4515 | 1.4524 | 0.2889 |
1.3552 | 22.0 | 4730 | 1.4519 | 0.2889 |
1.3869 | 23.0 | 4945 | 1.4514 | 0.2889 |
1.3741 | 24.0 | 5160 | 1.4510 | 0.2889 |
1.3721 | 25.0 | 5375 | 1.4505 | 0.2889 |
1.3593 | 26.0 | 5590 | 1.4501 | 0.2889 |
1.3536 | 27.0 | 5805 | 1.4497 | 0.2889 |
1.3543 | 28.0 | 6020 | 1.4493 | 0.2889 |
1.3589 | 29.0 | 6235 | 1.4489 | 0.2889 |
1.3445 | 30.0 | 6450 | 1.4486 | 0.2889 |
1.3539 | 31.0 | 6665 | 1.4483 | 0.2889 |
1.3535 | 32.0 | 6880 | 1.4480 | 0.2889 |
1.3498 | 33.0 | 7095 | 1.4477 | 0.2889 |
1.3497 | 34.0 | 7310 | 1.4474 | 0.2889 |
1.3582 | 35.0 | 7525 | 1.4472 | 0.2889 |
1.354 | 36.0 | 7740 | 1.4469 | 0.2889 |
1.3681 | 37.0 | 7955 | 1.4467 | 0.2889 |
1.346 | 38.0 | 8170 | 1.4465 | 0.2889 |
1.3468 | 39.0 | 8385 | 1.4463 | 0.2889 |
1.3488 | 40.0 | 8600 | 1.4462 | 0.2889 |
1.3542 | 41.0 | 8815 | 1.4460 | 0.2889 |
1.3813 | 42.0 | 9030 | 1.4459 | 0.2889 |
1.3585 | 43.0 | 9245 | 1.4458 | 0.2889 |
1.3347 | 44.0 | 9460 | 1.4457 | 0.2889 |
1.3527 | 45.0 | 9675 | 1.4456 | 0.2889 |
1.3601 | 46.0 | 9890 | 1.4456 | 0.2889 |
1.3484 | 47.0 | 10105 | 1.4455 | 0.2889 |
1.3543 | 48.0 | 10320 | 1.4455 | 0.2889 |
1.3639 | 49.0 | 10535 | 1.4455 | 0.2889 |
1.3697 | 50.0 | 10750 | 1.4455 | 0.2889 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.0+cu121
- Datasets 2.12.0
- Tokenizers 0.13.2
- Downloads last month
- 5