synergyai-jaeung's picture
Model save
953fe43 verified
|
raw
history blame
12.3 kB
metadata
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
  - f1
  - recall
  - precision
model-index:
  - name: vit-base-patch16-224-in21k_covid_19_ct_scans
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8887841658812441
          - name: F1
            type: f1
            value: 0.7572553125484722
          - name: Recall
            type: recall
            value: 0.9729119638826185
          - name: Precision
            type: precision
            value: 0.9016736401673641

vit-base-patch16-224-in21k_covid_19_ct_scans

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

  • Loss: 1.7287
  • Accuracy: 0.8888
  • F1: 0.7573
  • Auc: 0.7179
  • Recall: 0.9729
  • Precision: 0.9017

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: 0.0002
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 100

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Auc Recall Precision
0.768 1.0 266 0.4546 0.8351 0.4551 0.5 1.0 0.8351
0.4516 2.0 532 0.4498 0.8351 0.4551 0.5 1.0 0.8351
0.4516 3.0 798 0.4492 0.8351 0.4551 0.5 1.0 0.8351
0.4521 4.0 1064 0.4486 0.8351 0.4551 0.5 1.0 0.8351
0.4521 5.0 1330 0.4457 0.8351 0.4551 0.5 1.0 0.8351
0.4415 6.0 1596 0.4422 0.8351 0.4551 0.5 1.0 0.8351
0.4415 7.0 1862 0.4249 0.8351 0.4551 0.5 1.0 0.8351
0.4344 8.0 2128 0.4644 0.8351 0.4966 0.5183 0.9910 0.8402
0.4344 9.0 2394 0.4209 0.8407 0.5272 0.5355 0.9910 0.8450
0.3848 10.0 2660 0.4336 0.8030 0.6572 0.6642 0.8713 0.8904
0.3848 11.0 2926 0.4307 0.8407 0.6595 0.6387 0.9402 0.8778
0.2882 12.0 3192 0.5094 0.8219 0.6913 0.7007 0.8815 0.9029
0.2882 13.0 3458 0.4620 0.8520 0.6637 0.6363 0.9582 0.8762
0.1654 14.0 3724 0.5891 0.8351 0.7142 0.7247 0.8894 0.9110
0.1654 15.0 3990 0.5602 0.8417 0.6940 0.6828 0.9199 0.8936
0.0868 16.0 4256 0.5928 0.8690 0.7114 0.6785 0.9628 0.8895
0.045 17.0 4522 0.6154 0.8633 0.7268 0.7072 0.9402 0.9005
0.045 18.0 4788 0.6358 0.8680 0.7370 0.7169 0.9424 0.9037
0.021 19.0 5054 0.8247 0.8530 0.7379 0.7423 0.9074 0.9157
0.021 20.0 5320 0.9930 0.8473 0.7229 0.7229 0.9086 0.9086
0.0136 21.0 5586 0.5601 0.8652 0.7262 0.7038 0.9447 0.8990
0.0136 22.0 5852 0.6475 0.8699 0.6935 0.6562 0.9752 0.8816
0.0464 23.0 6118 0.5767 0.8567 0.7273 0.7170 0.9255 0.9051
0.0464 24.0 6384 0.7394 0.8501 0.7369 0.7452 0.9018 0.9173
0.0438 25.0 6650 0.7622 0.8680 0.6781 0.6413 0.9797 0.8768
0.0438 26.0 6916 0.7617 0.8831 0.7509 0.7168 0.9650 0.9019
0.0126 27.0 7182 0.8841 0.8624 0.7354 0.7227 0.9312 0.9066
0.0126 28.0 7448 0.7538 0.8784 0.7544 0.7300 0.9515 0.9074
0.016 29.0 7714 0.7106 0.8718 0.6709 0.6321 0.9898 0.8735
0.016 30.0 7980 0.6112 0.8756 0.7251 0.6893 0.9673 0.8927
0.0384 31.0 8246 0.5990 0.8784 0.7271 0.6887 0.9718 0.8922
0.0276 32.0 8512 0.6617 0.8850 0.7411 0.6996 0.9763 0.8954
0.0276 33.0 8778 0.7069 0.8907 0.7599 0.7190 0.9752 0.9019
0.0109 34.0 9044 0.8042 0.8746 0.6974 0.6567 0.9819 0.8815
0.0109 35.0 9310 0.7706 0.8831 0.7369 0.6962 0.9752 0.8944
0.0028 36.0 9576 0.8394 0.8869 0.7516 0.7122 0.9729 0.8998
0.0028 37.0 9842 0.8954 0.8850 0.7475 0.7087 0.9718 0.8987
0.0076 38.0 10108 0.9389 0.8850 0.7475 0.7087 0.9718 0.8987
0.0076 39.0 10374 0.9697 0.8850 0.7475 0.7087 0.9718 0.8987
0.0001 40.0 10640 0.9954 0.8850 0.7475 0.7087 0.9718 0.8987
0.0001 41.0 10906 1.0169 0.8850 0.7475 0.7087 0.9718 0.8987
0.0 42.0 11172 1.0381 0.8860 0.7488 0.7093 0.9729 0.8989
0.0 43.0 11438 1.0582 0.8860 0.7488 0.7093 0.9729 0.8989
0.0 44.0 11704 1.0763 0.8860 0.7488 0.7093 0.9729 0.8989
0.0 45.0 11970 1.0937 0.8860 0.7488 0.7093 0.9729 0.8989
0.0 46.0 12236 1.1095 0.8878 0.7545 0.7150 0.9729 0.9007
0.0 47.0 12502 1.1263 0.8878 0.7545 0.7150 0.9729 0.9007
0.0 48.0 12768 1.1427 0.8878 0.7545 0.7150 0.9729 0.9007
0.0 49.0 13034 1.1587 0.8878 0.7545 0.7150 0.9729 0.9007
0.0 50.0 13300 1.1745 0.8878 0.7545 0.7150 0.9729 0.9007
0.0 51.0 13566 1.1901 0.8878 0.7545 0.7150 0.9729 0.9007
0.0 52.0 13832 1.2052 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 53.0 14098 1.2201 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 54.0 14364 1.2350 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 55.0 14630 1.2497 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 56.0 14896 1.2641 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 57.0 15162 1.2785 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 58.0 15428 1.2925 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 59.0 15694 1.3068 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 60.0 15960 1.3207 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 61.0 16226 1.3346 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 62.0 16492 1.3485 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 63.0 16758 1.3622 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 64.0 17024 1.3758 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 65.0 17290 1.3893 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 66.0 17556 1.4029 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 67.0 17822 1.4166 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 68.0 18088 1.4298 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 69.0 18354 1.4431 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 70.0 18620 1.4566 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 71.0 18886 1.4695 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 72.0 19152 1.4824 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 73.0 19418 1.4950 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 74.0 19684 1.5076 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 75.0 19950 1.5201 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 76.0 20216 1.5321 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 77.0 20482 1.5441 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 78.0 20748 1.5564 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 79.0 21014 1.5691 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 80.0 21280 1.5800 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 81.0 21546 1.5910 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 82.0 21812 1.6021 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 83.0 22078 1.6133 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 84.0 22344 1.6244 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 85.0 22610 1.6357 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 86.0 22876 1.6468 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 87.0 23142 1.6580 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 88.0 23408 1.6694 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 89.0 23674 1.6806 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 90.0 23940 1.6876 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 91.0 24206 1.6938 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 92.0 24472 1.6996 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 93.0 24738 1.7051 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 94.0 25004 1.7104 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 95.0 25270 1.7152 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 96.0 25536 1.7195 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 97.0 25802 1.7232 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 98.0 26068 1.7260 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 99.0 26334 1.7280 0.8888 0.7573 0.7179 0.9729 0.9017
0.0 100.0 26600 1.7287 0.8888 0.7573 0.7179 0.9729 0.9017

Framework versions

  • Transformers 4.41.1
  • Pytorch 2.0.0+cu117
  • Datasets 2.19.1
  • Tokenizers 0.19.1