--- library_name: transformers license: apache-2.0 base_model: google/vit-base-patch16-224-in21k tags: - generated_from_trainer datasets: - imagefolder metrics: - accuracy - precision - recall - f1 model-index: - name: vit-base-kidney-stone-4-Michel_Daudon_-w256_1k_v1-_MIX 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: 0.8333333333333334 - name: Precision type: precision value: 0.8596272111598016 - name: Recall type: recall value: 0.8333333333333334 - name: F1 type: f1 value: 0.8312823840804323 --- # vit-base-kidney-stone-4-Michel_Daudon_-w256_1k_v1-_MIX This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset. It achieves the following results on the evaluation set: - Loss: 0.5183 - Accuracy: 0.8333 - Precision: 0.8596 - Recall: 0.8333 - F1: 0.8313 ## 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: 8 - seed: 42 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments - lr_scheduler_type: linear - num_epochs: 15 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | |:-------------:|:-------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:| | 0.4337 | 0.1667 | 100 | 0.6415 | 0.7688 | 0.7866 | 0.7688 | 0.7620 | | 0.5458 | 0.3333 | 200 | 1.0270 | 0.7204 | 0.8072 | 0.7204 | 0.6929 | | 0.1893 | 0.5 | 300 | 0.5183 | 0.8333 | 0.8596 | 0.8333 | 0.8313 | | 0.2041 | 0.6667 | 400 | 0.5611 | 0.8333 | 0.8651 | 0.8333 | 0.8360 | | 0.2087 | 0.8333 | 500 | 0.8036 | 0.7846 | 0.8253 | 0.7846 | 0.7916 | | 0.1888 | 1.0 | 600 | 0.7427 | 0.8046 | 0.8312 | 0.8046 | 0.7960 | | 0.1175 | 1.1667 | 700 | 0.7927 | 0.7837 | 0.7906 | 0.7837 | 0.7770 | | 0.5783 | 1.3333 | 800 | 0.9454 | 0.7521 | 0.8095 | 0.7521 | 0.7551 | | 0.1242 | 1.5 | 900 | 1.0772 | 0.7704 | 0.8102 | 0.7704 | 0.7796 | | 0.1045 | 1.6667 | 1000 | 0.8234 | 0.8296 | 0.8333 | 0.8296 | 0.8223 | | 0.1007 | 1.8333 | 1100 | 1.1756 | 0.7546 | 0.7483 | 0.7546 | 0.7460 | | 0.0101 | 2.0 | 1200 | 0.7921 | 0.8446 | 0.8782 | 0.8446 | 0.8486 | | 0.0079 | 2.1667 | 1300 | 0.9626 | 0.8204 | 0.8644 | 0.8204 | 0.8241 | | 0.0626 | 2.3333 | 1400 | 1.0140 | 0.8025 | 0.8441 | 0.8025 | 0.8040 | | 0.0216 | 2.5 | 1500 | 0.9297 | 0.8358 | 0.8540 | 0.8358 | 0.8364 | | 0.0707 | 2.6667 | 1600 | 0.9193 | 0.8196 | 0.8425 | 0.8196 | 0.8203 | | 0.0308 | 2.8333 | 1700 | 0.9988 | 0.8246 | 0.8429 | 0.8246 | 0.8209 | | 0.0863 | 3.0 | 1800 | 0.8083 | 0.83 | 0.8592 | 0.83 | 0.8332 | | 0.0016 | 3.1667 | 1900 | 1.1933 | 0.8029 | 0.8475 | 0.8029 | 0.8079 | | 0.0014 | 3.3333 | 2000 | 1.0995 | 0.8142 | 0.8376 | 0.8142 | 0.8132 | | 0.0745 | 3.5 | 2100 | 1.0348 | 0.8154 | 0.8720 | 0.8154 | 0.8259 | | 0.0226 | 3.6667 | 2200 | 0.8861 | 0.8275 | 0.8576 | 0.8275 | 0.8303 | | 0.0159 | 3.8333 | 2300 | 1.1476 | 0.79 | 0.8251 | 0.79 | 0.7981 | | 0.1398 | 4.0 | 2400 | 1.2559 | 0.7879 | 0.8284 | 0.7879 | 0.7845 | | 0.0011 | 4.1667 | 2500 | 1.2795 | 0.8008 | 0.8419 | 0.8008 | 0.8061 | | 0.0016 | 4.3333 | 2600 | 1.1345 | 0.8108 | 0.8472 | 0.8108 | 0.8154 | | 0.001 | 4.5 | 2700 | 1.0013 | 0.8242 | 0.8419 | 0.8242 | 0.8220 | | 0.0888 | 4.6667 | 2800 | 1.0708 | 0.8313 | 0.8614 | 0.8313 | 0.8357 | | 0.0212 | 4.8333 | 2900 | 1.1488 | 0.8113 | 0.8435 | 0.8113 | 0.8123 | | 0.0857 | 5.0 | 3000 | 1.0805 | 0.8113 | 0.8506 | 0.8113 | 0.8182 | | 0.0029 | 5.1667 | 3100 | 0.8731 | 0.8588 | 0.8762 | 0.8588 | 0.8619 | | 0.0226 | 5.3333 | 3200 | 1.2513 | 0.8113 | 0.8410 | 0.8113 | 0.8128 | | 0.0627 | 5.5 | 3300 | 1.1715 | 0.8063 | 0.8394 | 0.8063 | 0.8066 | | 0.1471 | 5.6667 | 3400 | 0.8260 | 0.8325 | 0.8434 | 0.8325 | 0.8341 | | 0.0008 | 5.8333 | 3500 | 0.8541 | 0.8404 | 0.8636 | 0.8404 | 0.8430 | | 0.0005 | 6.0 | 3600 | 1.1119 | 0.8129 | 0.8340 | 0.8129 | 0.8165 | | 0.0005 | 6.1667 | 3700 | 1.6586 | 0.7754 | 0.8261 | 0.7754 | 0.7762 | | 0.0693 | 6.3333 | 3800 | 1.2959 | 0.8067 | 0.8427 | 0.8067 | 0.8107 | | 0.0007 | 6.5 | 3900 | 1.0675 | 0.8142 | 0.8195 | 0.8142 | 0.8140 | | 0.0008 | 6.6667 | 4000 | 1.3692 | 0.7904 | 0.8078 | 0.7904 | 0.7903 | | 0.0063 | 6.8333 | 4100 | 1.2463 | 0.8092 | 0.8326 | 0.8092 | 0.8073 | | 0.0006 | 7.0 | 4200 | 1.2368 | 0.8171 | 0.8433 | 0.8171 | 0.8187 | | 0.0014 | 7.1667 | 4300 | 1.2245 | 0.7979 | 0.8126 | 0.7979 | 0.8004 | | 0.0005 | 7.3333 | 4400 | 1.2486 | 0.7996 | 0.8134 | 0.7996 | 0.7996 | | 0.0793 | 7.5 | 4500 | 1.3575 | 0.7762 | 0.8005 | 0.7762 | 0.7696 | | 0.0006 | 7.6667 | 4600 | 1.2693 | 0.8013 | 0.8151 | 0.8013 | 0.7996 | | 0.0005 | 7.8333 | 4700 | 1.1999 | 0.8192 | 0.8405 | 0.8192 | 0.8199 | | 0.0007 | 8.0 | 4800 | 1.0169 | 0.8346 | 0.8517 | 0.8346 | 0.8353 | | 0.067 | 8.1667 | 4900 | 1.0823 | 0.8346 | 0.8602 | 0.8346 | 0.8325 | | 0.0007 | 8.3333 | 5000 | 1.3014 | 0.7996 | 0.8439 | 0.7996 | 0.7978 | | 0.0003 | 8.5 | 5100 | 1.3176 | 0.7954 | 0.8398 | 0.7954 | 0.7986 | | 0.0003 | 8.6667 | 5200 | 1.2994 | 0.8113 | 0.8559 | 0.8113 | 0.8124 | | 0.0002 | 8.8333 | 5300 | 1.3460 | 0.7937 | 0.8308 | 0.7937 | 0.7908 | | 0.0003 | 9.0 | 5400 | 1.0408 | 0.8346 | 0.8541 | 0.8346 | 0.8363 | | 0.0002 | 9.1667 | 5500 | 1.1659 | 0.8246 | 0.8651 | 0.8246 | 0.8258 | | 0.0002 | 9.3333 | 5600 | 1.1821 | 0.8263 | 0.8657 | 0.8263 | 0.8270 | | 0.0002 | 9.5 | 5700 | 1.2786 | 0.8233 | 0.8607 | 0.8233 | 0.8227 | | 0.0002 | 9.6667 | 5800 | 1.2611 | 0.8217 | 0.8577 | 0.8217 | 0.8210 | | 0.0002 | 9.8333 | 5900 | 1.2556 | 0.8213 | 0.8568 | 0.8213 | 0.8206 | | 0.0002 | 10.0 | 6000 | 1.3472 | 0.8158 | 0.8491 | 0.8158 | 0.8158 | | 0.0002 | 10.1667 | 6100 | 1.3345 | 0.8175 | 0.8502 | 0.8175 | 0.8176 | | 0.0001 | 10.3333 | 6200 | 1.3366 | 0.8187 | 0.8512 | 0.8187 | 0.8188 | | 0.0001 | 10.5 | 6300 | 1.3363 | 0.8171 | 0.8497 | 0.8171 | 0.8174 | | 0.0001 | 10.6667 | 6400 | 1.3340 | 0.8196 | 0.8517 | 0.8196 | 0.8198 | | 0.0001 | 10.8333 | 6500 | 1.3658 | 0.8233 | 0.8593 | 0.8233 | 0.8243 | | 0.0001 | 11.0 | 6600 | 1.3709 | 0.8237 | 0.8595 | 0.8237 | 0.8247 | | 0.0001 | 11.1667 | 6700 | 1.3652 | 0.8242 | 0.8585 | 0.8242 | 0.8249 | | 0.0001 | 11.3333 | 6800 | 1.3703 | 0.825 | 0.8594 | 0.825 | 0.8258 | | 0.0001 | 11.5 | 6900 | 1.3755 | 0.8237 | 0.8579 | 0.8237 | 0.8247 | | 0.0001 | 11.6667 | 7000 | 1.3781 | 0.8237 | 0.8579 | 0.8237 | 0.8247 | | 0.0001 | 11.8333 | 7100 | 1.3811 | 0.8242 | 0.8582 | 0.8242 | 0.8251 | | 0.0001 | 12.0 | 7200 | 1.3851 | 0.8237 | 0.8578 | 0.8237 | 0.8247 | | 0.0001 | 12.1667 | 7300 | 1.3881 | 0.8242 | 0.8580 | 0.8242 | 0.8251 | | 0.0001 | 12.3333 | 7400 | 1.3910 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 12.5 | 7500 | 1.3937 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 12.6667 | 7600 | 1.3977 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 12.8333 | 7700 | 1.3995 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 13.0 | 7800 | 1.4021 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 13.1667 | 7900 | 1.4048 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 13.3333 | 8000 | 1.4074 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 13.5 | 8100 | 1.4099 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 13.6667 | 8200 | 1.4117 | 0.8246 | 0.8580 | 0.8246 | 0.8257 | | 0.0001 | 13.8333 | 8300 | 1.4134 | 0.825 | 0.8582 | 0.825 | 0.8261 | | 0.0001 | 14.0 | 8400 | 1.4150 | 0.825 | 0.8582 | 0.825 | 0.8261 | | 0.0001 | 14.1667 | 8500 | 1.4164 | 0.8246 | 0.8578 | 0.8246 | 0.8258 | | 0.0001 | 14.3333 | 8600 | 1.4176 | 0.8242 | 0.8574 | 0.8242 | 0.8254 | | 0.0001 | 14.5 | 8700 | 1.4186 | 0.8242 | 0.8574 | 0.8242 | 0.8254 | | 0.0001 | 14.6667 | 8800 | 1.4192 | 0.8242 | 0.8574 | 0.8242 | 0.8254 | | 0.0001 | 14.8333 | 8900 | 1.4197 | 0.8242 | 0.8574 | 0.8242 | 0.8254 | | 0.0001 | 15.0 | 9000 | 1.4200 | 0.8242 | 0.8574 | 0.8242 | 0.8254 | ### Framework versions - Transformers 4.48.2 - Pytorch 2.6.0+cu126 - Datasets 3.1.0 - Tokenizers 0.21.0