hkivancoral's picture
End of training
6a15123
metadata
license: apache-2.0
base_model: microsoft/beit-large-patch16-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: smids_10x_beit_large_sgd_00001_fold4
    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: 0.6483333333333333

smids_10x_beit_large_sgd_00001_fold4

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

  • Loss: 0.7900
  • Accuracy: 0.6483

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
1.2211 1.0 750 1.1931 0.3533
1.1323 2.0 1500 1.1612 0.365
1.184 3.0 2250 1.1336 0.37
1.1136 4.0 3000 1.1091 0.3817
0.9758 5.0 3750 1.0870 0.385
1.0842 6.0 4500 1.0669 0.3917
1.0165 7.0 5250 1.0484 0.4117
1.0062 8.0 6000 1.0310 0.43
1.0015 9.0 6750 1.0148 0.4367
0.9415 10.0 7500 0.9997 0.45
0.9588 11.0 8250 0.9856 0.4533
0.9674 12.0 9000 0.9724 0.47
0.9046 13.0 9750 0.9600 0.4733
0.9542 14.0 10500 0.9483 0.4867
0.8663 15.0 11250 0.9372 0.5
0.8717 16.0 12000 0.9268 0.51
0.7922 17.0 12750 0.9171 0.525
0.8562 18.0 13500 0.9078 0.535
0.9212 19.0 14250 0.8991 0.5433
0.8823 20.0 15000 0.8907 0.5567
0.8498 21.0 15750 0.8828 0.565
0.8335 22.0 16500 0.8754 0.575
0.8369 23.0 17250 0.8683 0.5867
0.8886 24.0 18000 0.8617 0.5917
0.8131 25.0 18750 0.8555 0.6
0.8107 26.0 19500 0.8497 0.605
0.7489 27.0 20250 0.8442 0.61
0.8154 28.0 21000 0.8390 0.6167
0.7935 29.0 21750 0.8341 0.62
0.7606 30.0 22500 0.8296 0.6267
0.7688 31.0 23250 0.8253 0.6283
0.755 32.0 24000 0.8214 0.63
0.8046 33.0 24750 0.8176 0.63
0.8193 34.0 25500 0.8142 0.6317
0.7668 35.0 26250 0.8110 0.635
0.7573 36.0 27000 0.8080 0.6367
0.7928 37.0 27750 0.8053 0.6417
0.792 38.0 28500 0.8028 0.6417
0.7917 39.0 29250 0.8007 0.645
0.7521 40.0 30000 0.7987 0.645
0.777 41.0 30750 0.7969 0.6483
0.7956 42.0 31500 0.7954 0.6483
0.8067 43.0 32250 0.7940 0.65
0.7335 44.0 33000 0.7929 0.65
0.7708 45.0 33750 0.7920 0.6483
0.74 46.0 34500 0.7912 0.6483
0.7222 47.0 35250 0.7906 0.6483
0.7572 48.0 36000 0.7902 0.6483
0.7909 49.0 36750 0.7900 0.6483
0.7055 50.0 37500 0.7900 0.6483

Framework versions

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