hkivancoral's picture
End of training
ce0e616
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_fold2
    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: 0.6888888888888889

hushem_40x_deit_tiny_sgd_001_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: 1.0440
  • 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: 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
1.0975 1.0 215 1.3370 0.3778
0.8761 2.0 430 1.2895 0.4444
0.7359 3.0 645 1.2565 0.4889
0.6277 4.0 860 1.2398 0.5556
0.5094 5.0 1075 1.2052 0.5556
0.4187 6.0 1290 1.1950 0.5778
0.3909 7.0 1505 1.1310 0.6
0.3137 8.0 1720 1.1412 0.5556
0.2817 9.0 1935 1.0706 0.5778
0.2108 10.0 2150 1.0537 0.6
0.1785 11.0 2365 1.0606 0.5778
0.1677 12.0 2580 1.0202 0.5778
0.1602 13.0 2795 1.0251 0.5778
0.1355 14.0 3010 1.0164 0.6
0.1234 15.0 3225 1.0019 0.5778
0.0937 16.0 3440 0.9960 0.6
0.0963 17.0 3655 0.9708 0.5778
0.0998 18.0 3870 0.9907 0.5778
0.0604 19.0 4085 0.9932 0.6
0.0724 20.0 4300 0.9792 0.5556
0.0616 21.0 4515 0.9528 0.5556
0.0591 22.0 4730 0.9741 0.5556
0.0433 23.0 4945 0.9824 0.5556
0.0476 24.0 5160 0.9907 0.5556
0.0326 25.0 5375 0.9714 0.5778
0.0325 26.0 5590 0.9834 0.6
0.0352 27.0 5805 0.9903 0.5778
0.0319 28.0 6020 0.9831 0.5778
0.0242 29.0 6235 0.9872 0.6
0.0238 30.0 6450 1.0027 0.6222
0.0166 31.0 6665 0.9985 0.5778
0.0151 32.0 6880 1.0088 0.6
0.0176 33.0 7095 1.0180 0.6
0.0221 34.0 7310 1.0038 0.6444
0.0159 35.0 7525 0.9868 0.6667
0.0115 36.0 7740 1.0104 0.6444
0.017 37.0 7955 1.0128 0.6889
0.0105 38.0 8170 1.0250 0.6444
0.0144 39.0 8385 1.0115 0.6889
0.0092 40.0 8600 1.0202 0.6667
0.0131 41.0 8815 1.0296 0.6444
0.0108 42.0 9030 1.0274 0.6889
0.0089 43.0 9245 1.0423 0.6889
0.0153 44.0 9460 1.0420 0.6889
0.0077 45.0 9675 1.0387 0.6667
0.0096 46.0 9890 1.0413 0.6889
0.0073 47.0 10105 1.0431 0.6889
0.0112 48.0 10320 1.0453 0.6889
0.0085 49.0 10535 1.0438 0.6889
0.01 50.0 10750 1.0440 0.6889

Framework versions

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