hkivancoral's picture
End of training
188971e
metadata
license: apache-2.0
base_model: facebook/deit-tiny-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_5x_deit_tiny_sgd_00001_fold1
    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: 0.4540901502504174

smids_5x_deit_tiny_sgd_00001_fold1

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: 1.0635
  • Accuracy: 0.4541

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
1.346 1.0 376 1.2991 0.3489
1.3817 2.0 752 1.2686 0.3589
1.3103 3.0 1128 1.2425 0.3656
1.3556 4.0 1504 1.2205 0.3656
1.2443 5.0 1880 1.2020 0.3723
1.1947 6.0 2256 1.1865 0.3806
1.184 7.0 2632 1.1737 0.3940
1.2121 8.0 3008 1.1630 0.3873
1.1793 9.0 3384 1.1540 0.3773
1.1564 10.0 3760 1.1464 0.3740
1.148 11.0 4136 1.1397 0.3756
1.1774 12.0 4512 1.1340 0.3756
1.1493 13.0 4888 1.1288 0.3790
1.1491 14.0 5264 1.1241 0.3790
1.1465 15.0 5640 1.1198 0.3856
1.1089 16.0 6016 1.1159 0.3990
1.1015 17.0 6392 1.1122 0.4057
1.1166 18.0 6768 1.1086 0.4073
1.1502 19.0 7144 1.1053 0.4124
1.124 20.0 7520 1.1022 0.4174
1.1102 21.0 7896 1.0992 0.4207
1.0904 22.0 8272 1.0964 0.4190
1.0897 23.0 8648 1.0937 0.4207
1.1449 24.0 9024 1.0912 0.4190
1.0609 25.0 9400 1.0888 0.4157
1.0747 26.0 9776 1.0865 0.4207
1.0631 27.0 10152 1.0844 0.4240
1.0872 28.0 10528 1.0823 0.4274
1.0811 29.0 10904 1.0804 0.4290
1.1082 30.0 11280 1.0786 0.4307
1.0863 31.0 11656 1.0769 0.4324
1.103 32.0 12032 1.0753 0.4290
1.0918 33.0 12408 1.0738 0.4324
1.06 34.0 12784 1.0725 0.4391
1.0723 35.0 13160 1.0712 0.4424
1.0366 36.0 13536 1.0701 0.4457
1.0655 37.0 13912 1.0690 0.4474
1.0787 38.0 14288 1.0681 0.4457
1.0751 39.0 14664 1.0672 0.4474
1.0508 40.0 15040 1.0665 0.4541
1.0565 41.0 15416 1.0658 0.4541
1.0404 42.0 15792 1.0652 0.4541
1.0767 43.0 16168 1.0648 0.4541
1.076 44.0 16544 1.0644 0.4541
1.0183 45.0 16920 1.0640 0.4541
1.0393 46.0 17296 1.0638 0.4541
1.065 47.0 17672 1.0636 0.4541
1.0432 48.0 18048 1.0635 0.4541
1.0432 49.0 18424 1.0635 0.4541
1.0255 50.0 18800 1.0635 0.4541

Framework versions

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