Edit model card

hushem_5x_beit_base_adamax_00001_fold1

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

  • Loss: 1.3063
  • 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: 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
1.2914 1.0 27 1.3981 0.3111
0.7752 2.0 54 1.2587 0.4222
0.473 3.0 81 1.1213 0.6
0.3517 4.0 108 1.0654 0.5778
0.2036 5.0 135 0.9700 0.6222
0.1396 6.0 162 0.9127 0.6444
0.1055 7.0 189 1.0554 0.6222
0.0683 8.0 216 0.9132 0.6222
0.0509 9.0 243 1.0907 0.6222
0.0285 10.0 270 1.0220 0.6667
0.0302 11.0 297 0.9814 0.6667
0.0178 12.0 324 1.0288 0.6667
0.0215 13.0 351 0.9906 0.6667
0.0098 14.0 378 0.9906 0.6667
0.0094 15.0 405 0.9909 0.6667
0.0079 16.0 432 1.0583 0.6889
0.0176 17.0 459 1.0002 0.7111
0.0071 18.0 486 1.1076 0.7111
0.0077 19.0 513 1.2658 0.7111
0.0085 20.0 540 1.2202 0.7111
0.0042 21.0 567 1.1485 0.6889
0.0109 22.0 594 1.1833 0.6667
0.0017 23.0 621 1.2496 0.6667
0.0025 24.0 648 1.2268 0.6667
0.0049 25.0 675 1.1304 0.6889
0.0023 26.0 702 1.0752 0.6667
0.002 27.0 729 1.3029 0.6889
0.0019 28.0 756 1.1867 0.6444
0.0014 29.0 783 1.1802 0.7333
0.002 30.0 810 1.3660 0.7111
0.0126 31.0 837 1.3022 0.6889
0.002 32.0 864 1.3902 0.6667
0.0046 33.0 891 1.3937 0.6889
0.0019 34.0 918 1.3856 0.7333
0.0048 35.0 945 1.3752 0.6667
0.002 36.0 972 1.3963 0.6667
0.0009 37.0 999 1.3895 0.7111
0.001 38.0 1026 1.2536 0.6889
0.0016 39.0 1053 1.2991 0.6667
0.0008 40.0 1080 1.2492 0.6889
0.0031 41.0 1107 1.2808 0.6889
0.0025 42.0 1134 1.3015 0.6667
0.0032 43.0 1161 1.3785 0.7111
0.0006 44.0 1188 1.3466 0.6889
0.004 45.0 1215 1.3569 0.6667
0.0022 46.0 1242 1.3406 0.6444
0.0027 47.0 1269 1.3081 0.6889
0.0015 48.0 1296 1.3063 0.6889
0.002 49.0 1323 1.3063 0.6889
0.003 50.0 1350 1.3063 0.6889

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu118
  • Datasets 2.15.0
  • Tokenizers 0.15.0
Downloads last month
1
Safetensors
Model size
85.8M params
Tensor type
F32
·
Inference API
Drag image file here or click to browse from your device
This model can be loaded on Inference API (serverless).

Finetuned from

Evaluation results