--- license: apache-2.0 tags: - generated_from_trainer metrics: - accuracy - f1 - precision - recall model-index: - name: vit-base-skin results: [] --- # vit-base-skin This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.6917 - Accuracy: 0.8549 - F1: 0.8552 - Precision: 0.8560 - Recall: 0.8549 ## 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: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 6 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:| | 0.9322 | 0.16 | 100 | 0.8109 | 0.6943 | 0.6290 | 0.5939 | 0.6943 | | 0.7518 | 0.32 | 200 | 0.6722 | 0.7409 | 0.6832 | 0.6945 | 0.7409 | | 0.6616 | 0.48 | 300 | 0.7126 | 0.7358 | 0.7077 | 0.7039 | 0.7358 | | 0.8264 | 0.64 | 400 | 0.6001 | 0.8135 | 0.8092 | 0.8178 | 0.8135 | | 0.5767 | 0.8 | 500 | 0.6306 | 0.7772 | 0.7619 | 0.7945 | 0.7772 | | 0.5939 | 0.96 | 600 | 0.4621 | 0.8290 | 0.7988 | 0.8397 | 0.8290 | | 0.4351 | 1.12 | 700 | 0.5544 | 0.7979 | 0.7894 | 0.8410 | 0.7979 | | 0.4737 | 1.28 | 800 | 0.5151 | 0.8238 | 0.8334 | 0.8708 | 0.8238 | | 0.428 | 1.44 | 900 | 0.4980 | 0.8238 | 0.8170 | 0.8299 | 0.8238 | | 0.4596 | 1.6 | 1000 | 0.5650 | 0.7927 | 0.8032 | 0.8428 | 0.7927 | | 0.4096 | 1.76 | 1100 | 0.4544 | 0.8342 | 0.8178 | 0.8567 | 0.8342 | | 0.4328 | 1.92 | 1200 | 0.4524 | 0.8290 | 0.8294 | 0.8482 | 0.8290 | | 0.2272 | 2.08 | 1300 | 0.4808 | 0.8290 | 0.8304 | 0.8409 | 0.8290 | | 0.2415 | 2.24 | 1400 | 0.5585 | 0.7927 | 0.7916 | 0.8057 | 0.7927 | | 0.2743 | 2.4 | 1500 | 0.4144 | 0.8497 | 0.8484 | 0.8497 | 0.8497 | | 0.1943 | 2.56 | 1600 | 0.3977 | 0.8705 | 0.8722 | 0.8761 | 0.8705 | | 0.1839 | 2.72 | 1700 | 0.4784 | 0.8394 | 0.8382 | 0.8517 | 0.8394 | | 0.1905 | 2.88 | 1800 | 0.4314 | 0.8653 | 0.8669 | 0.8724 | 0.8653 | | 0.111 | 3.04 | 1900 | 0.5080 | 0.8290 | 0.8309 | 0.8348 | 0.8290 | | 0.0872 | 3.19 | 2000 | 0.5320 | 0.8549 | 0.8520 | 0.8649 | 0.8549 | | 0.1169 | 3.35 | 2100 | 0.5110 | 0.8342 | 0.8386 | 0.8477 | 0.8342 | | 0.1181 | 3.51 | 2200 | 0.4916 | 0.8446 | 0.8482 | 0.8563 | 0.8446 | | 0.0879 | 3.67 | 2300 | 0.5428 | 0.8601 | 0.8657 | 0.8829 | 0.8601 | | 0.1896 | 3.83 | 2400 | 0.5534 | 0.8497 | 0.8484 | 0.8536 | 0.8497 | | 0.0794 | 3.99 | 2500 | 0.6542 | 0.8342 | 0.8259 | 0.8270 | 0.8342 | | 0.0398 | 4.15 | 2600 | 0.5962 | 0.8187 | 0.8243 | 0.8338 | 0.8187 | | 0.0512 | 4.31 | 2700 | 0.6286 | 0.8497 | 0.8447 | 0.8457 | 0.8497 | | 0.0106 | 4.47 | 2800 | 0.6446 | 0.8394 | 0.8372 | 0.8377 | 0.8394 | | 0.0058 | 4.63 | 2900 | 0.5754 | 0.8653 | 0.8616 | 0.8618 | 0.8653 | | 0.0268 | 4.79 | 3000 | 0.5966 | 0.8653 | 0.8651 | 0.8658 | 0.8653 | | 0.0146 | 4.95 | 3100 | 0.6707 | 0.8601 | 0.8535 | 0.8577 | 0.8601 | | 0.0325 | 5.11 | 3200 | 0.6543 | 0.8549 | 0.8518 | 0.8511 | 0.8549 | | 0.0063 | 5.27 | 3300 | 0.6780 | 0.8497 | 0.8519 | 0.8583 | 0.8497 | | 0.003 | 5.43 | 3400 | 0.6675 | 0.8601 | 0.8577 | 0.8562 | 0.8601 | | 0.0143 | 5.59 | 3500 | 0.6967 | 0.8601 | 0.8554 | 0.8539 | 0.8601 | | 0.004 | 5.75 | 3600 | 0.6992 | 0.8601 | 0.8573 | 0.8552 | 0.8601 | | 0.003 | 5.91 | 3700 | 0.6917 | 0.8549 | 0.8552 | 0.8560 | 0.8549 | ### Framework versions - Transformers 4.29.2 - Pytorch 1.13.1 - Datasets 2.14.5 - Tokenizers 0.13.3