Edit model card

smids_5x_deit_base_rms_00001_fold2

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

  • Loss: 1.0612
  • Accuracy: 0.8852

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
0.2146 1.0 375 0.2894 0.8952
0.1437 2.0 750 0.3102 0.8935
0.0617 3.0 1125 0.4639 0.8902
0.0286 4.0 1500 0.5331 0.8935
0.0028 5.0 1875 0.6331 0.8802
0.0027 6.0 2250 0.6763 0.8902
0.0012 7.0 2625 0.7498 0.8835
0.0002 8.0 3000 0.6659 0.8952
0.0 9.0 3375 0.7256 0.8935
0.0209 10.0 3750 0.7936 0.8769
0.0042 11.0 4125 0.8361 0.8802
0.001 12.0 4500 0.8162 0.8852
0.0011 13.0 4875 0.8014 0.8968
0.0 14.0 5250 0.8392 0.8885
0.0018 15.0 5625 0.9229 0.8752
0.0 16.0 6000 0.8989 0.8869
0.0029 17.0 6375 0.8923 0.8902
0.0 18.0 6750 0.8680 0.8852
0.0112 19.0 7125 0.9026 0.8852
0.0 20.0 7500 0.9170 0.8902
0.0 21.0 7875 1.0300 0.8735
0.0 22.0 8250 0.8953 0.8885
0.0 23.0 8625 0.9292 0.8918
0.0001 24.0 9000 0.9442 0.8935
0.0 25.0 9375 0.9984 0.8952
0.0 26.0 9750 1.0751 0.8885
0.0 27.0 10125 1.0903 0.8819
0.0 28.0 10500 1.0301 0.8852
0.0 29.0 10875 1.0019 0.8885
0.0 30.0 11250 0.9825 0.8902
0.0045 31.0 11625 1.0018 0.8835
0.0032 32.0 12000 1.0070 0.8885
0.0037 33.0 12375 0.9955 0.8902
0.0 34.0 12750 1.0401 0.8802
0.0 35.0 13125 1.0361 0.8835
0.0 36.0 13500 1.0263 0.8869
0.0 37.0 13875 1.0646 0.8802
0.0 38.0 14250 1.0823 0.8835
0.0 39.0 14625 1.0786 0.8852
0.0029 40.0 15000 1.0585 0.8869
0.0 41.0 15375 1.0567 0.8852
0.0023 42.0 15750 1.0631 0.8852
0.0027 43.0 16125 1.0573 0.8885
0.0026 44.0 16500 1.0579 0.8902
0.0023 45.0 16875 1.0642 0.8852
0.0 46.0 17250 1.0620 0.8852
0.0049 47.0 17625 1.0628 0.8852
0.0 48.0 18000 1.0622 0.8852
0.0022 49.0 18375 1.0616 0.8852
0.0024 50.0 18750 1.0612 0.8852

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.1.0+cu121
  • Datasets 2.12.0
  • Tokenizers 0.13.2
Downloads last month
14

Finetuned from

Evaluation results