深度学习 Unet+Resnet 实战分割项目、多尺度训练、多类别分割：BraTS3d脑肿瘤图像切分的2D图片分割项目（4分类）

[train hyper-parameters: Namespace(base_size=400, batch_size=16, crop_size=256, epochs=10, lr=0.01, lrf=0.001)] [epoch: 1] global correct: 0.9716 precision: ['0.9934', '0.6329', '0.3479', '0.3796'] recall: ['0.9934', '0.7185', '0.3998', '0.1878'] IoU: ['0.9868', '0.5072', '0.2286', '0.1437'] mean IoU: 0.4666 [epoch: 2] global correct: 0.9758 precision: ['0.9934', '0.6556', '0.5929', '0.4193'] recall: ['0.9965', '0.7658', '0.1590', '0.3923'] IoU: ['0.9900', '0.5461', '0.1434', '0.2542'] mean IoU: 0.4834 [epoch: 3] global correct: 0.9755 precision: ['0.9894', '0.6570', '0.6022', '0.5040'] recall: ['0.9988', '0.7176', '0.1980', '0.2540'] IoU: ['0.9882', '0.5220', '0.1751', '0.2032'] mean IoU: 0.4721 [epoch: 4] global correct: 0.9775 precision: ['0.9938', '0.6901', '0.6620', '0.4663'] recall: ['0.9968', '0.7634', '0.1813', '0.5161'] IoU: ['0.9906', '0.5685', '0.1659', '0.3244'] mean IoU: 0.5124 [epoch: 5] global correct: 0.9788 precision: ['0.9949', '0.7022', '0.6789', '0.4913'] recall: ['0.9966', '0.8055', '0.1941', '0.5448'] IoU: ['0.9915', '0.6004', '0.1778', '0.3483'] mean IoU: 0.5295 [epoch: 6] global correct: 0.9789 precision: ['0.9924', '0.7266', '0.6647', '0.5247'] recall: ['0.9982', '0.7564', '0.2226', '0.5098'] IoU: ['0.9907', '0.5888', '0.2002', '0.3488'] mean IoU: 0.5321 [epoch: 7] global correct: 0.9801 precision: ['0.9938', '0.7817', '0.6202', '0.5068'] recall: ['0.9980', '0.7256', '0.3424', '0.6149'] IoU: ['0.9918', '0.6033', '0.2830', '0.3847'] mean IoU: 0.5657 [epoch: 8] global correct: 0.9804 precision: ['0.9942', '0.7829', '0.5746', '0.5233'] recall: ['0.9980', '0.7266', '0.4267', '0.5661'] IoU: ['0.9922', '0.6047', '0.3243', '0.3735'] mean IoU: 0.5737 [epoch: 9] global correct: 0.9815 precision: ['0.9948', '0.7471', '0.6227', '0.5930'] recall: ['0.9978', '0.8016', '0.4029', '0.5090'] IoU: ['0.9926', '0.6305', '0.3238', '0.3772'] mean IoU: 0.5811 [epoch: 10] global correct: 0.9817 precision: ['0.9950', '0.7682', '0.6219', '0.5668'] recall: ['0.9977', '0.7849', '0.4018', '0.5781'] IoU: ['0.9927', '0.6346', '0.3229', '0.4009'] mean IoU: 0.5878