Preliminary benchmarking of different CNN backbones on the dataset. Results report the average balanced accuracy across five cross-validation folds with 10 hyperparameter tuning trials and without fine-tuning. Among the tested architectures, ResNet-50 achieved the best performance (81.90 %).
收藏NIAID Data Ecosystem2026-05-10 收录
下载链接：
https://figshare.com/articles/dataset/Preliminary_benchmarking_of_different_CNN_backbones_on_the_dataset_Results_report_the_average_balanced_accuracy_across_five_cross-validation_folds_with_10_hyperparameter_tuning_trials_and_without_fine-tuning_Among_the_tested_architectures_R/30558796
下载链接
链接失效反馈官方服务：
资源简介：
Preliminary benchmarking of different CNN backbones on the dataset. Results report the average balanced accuracy across five cross-validation folds with 10 hyperparameter tuning trials and without fine-tuning. Among the tested architectures, ResNet-50 achieved the best performance (81.90 %).
创建时间：
2025-11-06
