smids_1x_deit_tiny_adamax_00001_fold4

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: 0.9763
  • Accuracy: 0.86

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
0.7004 1.0 75 0.6502 0.7417
0.5011 2.0 150 0.4927 0.795
0.387 3.0 225 0.4418 0.8133
0.32 4.0 300 0.4153 0.835
0.2817 5.0 375 0.4075 0.8333
0.2233 6.0 450 0.3963 0.8433
0.2099 7.0 525 0.3993 0.8517
0.1587 8.0 600 0.4089 0.8517
0.1492 9.0 675 0.4263 0.8533
0.1374 10.0 750 0.4223 0.86
0.1106 11.0 825 0.4566 0.8533
0.0777 12.0 900 0.4729 0.8433
0.0787 13.0 975 0.4848 0.8567
0.0407 14.0 1050 0.5231 0.85
0.0237 15.0 1125 0.5607 0.8533
0.0206 16.0 1200 0.6045 0.8483
0.0395 17.0 1275 0.6305 0.855
0.0399 18.0 1350 0.6575 0.85
0.0183 19.0 1425 0.7037 0.8467
0.0153 20.0 1500 0.7259 0.855
0.005 21.0 1575 0.7531 0.85
0.0017 22.0 1650 0.7833 0.85
0.0025 23.0 1725 0.8056 0.8467
0.0157 24.0 1800 0.8300 0.855
0.007 25.0 1875 0.8590 0.85
0.0006 26.0 1950 0.8573 0.8483
0.0006 27.0 2025 0.8637 0.8483
0.0055 28.0 2100 0.8671 0.8533
0.0223 29.0 2175 0.8791 0.85
0.0007 30.0 2250 0.8908 0.845
0.0004 31.0 2325 0.8875 0.85
0.0003 32.0 2400 0.8993 0.8533
0.0004 33.0 2475 0.9065 0.8517
0.0002 34.0 2550 0.9107 0.855
0.0003 35.0 2625 0.9197 0.855
0.0002 36.0 2700 0.9278 0.8517
0.02 37.0 2775 0.9327 0.8633
0.0002 38.0 2850 0.9380 0.8617
0.0002 39.0 2925 0.9441 0.8567
0.0003 40.0 3000 0.9479 0.86
0.0203 41.0 3075 0.9563 0.8583
0.0002 42.0 3150 0.9601 0.86
0.0002 43.0 3225 0.9632 0.86
0.0002 44.0 3300 0.9684 0.8533
0.0002 45.0 3375 0.9703 0.86
0.0002 46.0 3450 0.9720 0.86
0.0002 47.0 3525 0.9736 0.86
0.0002 48.0 3600 0.9753 0.86
0.0001 49.0 3675 0.9759 0.86
0.0001 50.0 3750 0.9763 0.86

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu118
  • Datasets 2.15.0
  • Tokenizers 0.15.0
Downloads last month
20
Safetensors
Model size
5.52M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and the model is not deployed on the HF Inference API.

Model tree for hkivancoral/smids_1x_deit_tiny_adamax_00001_fold4

Finetuned
(353)
this model

Evaluation results