hkivancoral's picture
End of training
9577825
metadata
license: apache-2.0
base_model: facebook/deit-small-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_3x_deit_small_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: 0.8569051580698835

smids_3x_deit_small_sgd_001_fold2

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3562
  • Accuracy: 0.8569

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
0.8599 1.0 225 0.8308 0.6789
0.6572 2.0 450 0.6508 0.7488
0.5628 3.0 675 0.5574 0.7920
0.4888 4.0 900 0.5053 0.8053
0.4209 5.0 1125 0.4713 0.8136
0.3729 6.0 1350 0.4472 0.8253
0.4088 7.0 1575 0.4292 0.8270
0.3871 8.0 1800 0.4168 0.8353
0.3605 9.0 2025 0.4049 0.8369
0.2963 10.0 2250 0.3978 0.8403
0.3519 11.0 2475 0.3893 0.8469
0.3126 12.0 2700 0.3814 0.8536
0.2889 13.0 2925 0.3781 0.8519
0.3096 14.0 3150 0.3739 0.8552
0.3153 15.0 3375 0.3694 0.8586
0.3271 16.0 3600 0.3680 0.8619
0.2697 17.0 3825 0.3655 0.8602
0.2138 18.0 4050 0.3624 0.8602
0.2422 19.0 4275 0.3602 0.8636
0.288 20.0 4500 0.3609 0.8652
0.3039 21.0 4725 0.3587 0.8619
0.2907 22.0 4950 0.3580 0.8619
0.3138 23.0 5175 0.3576 0.8652
0.2718 24.0 5400 0.3569 0.8669
0.263 25.0 5625 0.3551 0.8669
0.245 26.0 5850 0.3538 0.8686
0.2019 27.0 6075 0.3530 0.8636
0.2353 28.0 6300 0.3539 0.8636
0.2451 29.0 6525 0.3558 0.8602
0.2565 30.0 6750 0.3536 0.8652
0.2202 31.0 6975 0.3542 0.8636
0.2433 32.0 7200 0.3552 0.8636
0.2621 33.0 7425 0.3534 0.8652
0.2353 34.0 7650 0.3541 0.8652
0.1836 35.0 7875 0.3533 0.8602
0.2199 36.0 8100 0.3554 0.8619
0.2271 37.0 8325 0.3536 0.8602
0.1937 38.0 8550 0.3541 0.8619
0.1782 39.0 8775 0.3547 0.8586
0.1988 40.0 9000 0.3551 0.8586
0.1613 41.0 9225 0.3546 0.8602
0.1997 42.0 9450 0.3550 0.8586
0.1938 43.0 9675 0.3554 0.8569
0.1972 44.0 9900 0.3557 0.8586
0.2519 45.0 10125 0.3555 0.8569
0.2003 46.0 10350 0.3557 0.8569
0.1846 47.0 10575 0.3560 0.8586
0.1808 48.0 10800 0.3561 0.8569
0.2273 49.0 11025 0.3561 0.8569
0.1583 50.0 11250 0.3562 0.8569

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.1.0+cu121
  • Datasets 2.12.0
  • Tokenizers 0.13.2