Edit model card

smids_5x_deit_tiny_sgd_00001_fold5

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: 1.0692
  • Accuracy: 0.45

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
1.4013 1.0 375 1.3267 0.3533
1.3461 2.0 750 1.2947 0.36
1.2972 3.0 1125 1.2670 0.3617
1.31 4.0 1500 1.2433 0.3683
1.2221 5.0 1875 1.2236 0.3717
1.2656 6.0 2250 1.2067 0.375
1.2312 7.0 2625 1.1923 0.3817
1.1861 8.0 3000 1.1803 0.3817
1.1289 9.0 3375 1.1699 0.385
1.218 10.0 3750 1.1612 0.3867
1.1921 11.0 4125 1.1535 0.3983
1.1315 12.0 4500 1.1468 0.405
1.1732 13.0 4875 1.1407 0.4133
1.1412 14.0 5250 1.1354 0.41
1.1502 15.0 5625 1.1305 0.4133
1.126 16.0 6000 1.1259 0.4133
1.1098 17.0 6375 1.1217 0.4117
1.1197 18.0 6750 1.1177 0.4083
1.1329 19.0 7125 1.1140 0.4083
1.0741 20.0 7500 1.1105 0.4117
1.0617 21.0 7875 1.1072 0.4117
1.0917 22.0 8250 1.1041 0.41
1.0822 23.0 8625 1.1011 0.41
1.1336 24.0 9000 1.0984 0.4167
1.0665 25.0 9375 1.0958 0.415
1.1097 26.0 9750 1.0934 0.415
1.0499 27.0 10125 1.0911 0.4183
1.1202 28.0 10500 1.0889 0.4167
1.1038 29.0 10875 1.0869 0.4283
1.0838 30.0 11250 1.0850 0.4333
1.0717 31.0 11625 1.0832 0.4367
1.0773 32.0 12000 1.0816 0.4383
1.0858 33.0 12375 1.0800 0.44
1.0072 34.0 12750 1.0786 0.44
1.0435 35.0 13125 1.0773 0.4417
1.047 36.0 13500 1.0761 0.4433
1.0361 37.0 13875 1.0750 0.4483
1.0477 38.0 14250 1.0740 0.45
1.0658 39.0 14625 1.0731 0.4483
1.0711 40.0 15000 1.0723 0.445
1.0473 41.0 15375 1.0716 0.445
1.0521 42.0 15750 1.0711 0.445
1.0368 43.0 16125 1.0705 0.4467
1.0636 44.0 16500 1.0701 0.4483
1.0424 45.0 16875 1.0698 0.4483
1.0442 46.0 17250 1.0695 0.45
1.0667 47.0 17625 1.0694 0.45
1.0523 48.0 18000 1.0693 0.45
1.0135 49.0 18375 1.0692 0.45
1.0393 50.0 18750 1.0692 0.45

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.1.1+cu121
  • Datasets 2.12.0
  • Tokenizers 0.13.2
Downloads last month
14

Finetuned from

Evaluation results