Hugging Face's logo

hkivancoral
/
hushem_40x_deit_tiny_sgd_001_fold3

Image Classification
Transformers
PyTorch
vit
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
hushem_40x_deit_tiny_sgd_001_fold3 / README.md
hkivancoral's picture
hkivancoral
End of training
fa52a87
preview code
|
raw
history blame contribute delete
4.87 kB
metadata
license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: hushem_40x_deit_tiny_sgd_001_fold3
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8372093023255814

hushem_40x_deit_tiny_sgd_001_fold3

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: 0.4899
  • Accuracy: 0.8372

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: 0.001
  • 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
1.1798 1.0 217 1.2889 0.4186
1.0098 2.0 434 1.1067 0.6047
0.7827 3.0 651 0.9427 0.6977
0.6326 4.0 868 0.7917 0.6977
0.5443 5.0 1085 0.6647 0.7907
0.4438 6.0 1302 0.5810 0.8140
0.3761 7.0 1519 0.5185 0.8372
0.3386 8.0 1736 0.4778 0.8140
0.2796 9.0 1953 0.4431 0.8605
0.2037 10.0 2170 0.4372 0.8605
0.1624 11.0 2387 0.3943 0.8837
0.1477 12.0 2604 0.4019 0.8605
0.1485 13.0 2821 0.3856 0.8605
0.1192 14.0 3038 0.3686 0.8605
0.1115 15.0 3255 0.3722 0.8605
0.0891 16.0 3472 0.3567 0.8837
0.0776 17.0 3689 0.3631 0.8605
0.1039 18.0 3906 0.3600 0.8605
0.0608 19.0 4123 0.3514 0.8605
0.0639 20.0 4340 0.3706 0.8605
0.0555 21.0 4557 0.3773 0.8605
0.0552 22.0 4774 0.3713 0.8372
0.0457 23.0 4991 0.3749 0.8372
0.0383 24.0 5208 0.3901 0.8372
0.0332 25.0 5425 0.3933 0.8372
0.0322 26.0 5642 0.3995 0.8372
0.0278 27.0 5859 0.4012 0.8372
0.0212 28.0 6076 0.3938 0.8372
0.0224 29.0 6293 0.4080 0.8372
0.0218 30.0 6510 0.4237 0.8372
0.0278 31.0 6727 0.4231 0.8372
0.0212 32.0 6944 0.4330 0.8372
0.021 33.0 7161 0.4507 0.8372
0.0127 34.0 7378 0.4390 0.8372
0.0158 35.0 7595 0.4566 0.8372
0.0178 36.0 7812 0.4594 0.8372
0.0109 37.0 8029 0.4570 0.8372
0.0096 38.0 8246 0.4635 0.8372
0.0113 39.0 8463 0.4700 0.8372
0.0149 40.0 8680 0.4815 0.8372
0.0111 41.0 8897 0.4769 0.8372
0.0075 42.0 9114 0.4756 0.8372
0.0093 43.0 9331 0.4800 0.8372
0.009 44.0 9548 0.4851 0.8372
0.0065 45.0 9765 0.4808 0.8372
0.011 46.0 9982 0.4835 0.8372
0.0064 47.0 10199 0.4871 0.8372
0.0093 48.0 10416 0.4902 0.8372
0.0136 49.0 10633 0.4899 0.8372
0.0058 50.0 10850 0.4899 0.8372

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.1.1+cu121
  • Datasets 2.12.0
  • Tokenizers 0.13.2