ACNE04 读数据的过程

 交叉验证

cross_val_lists = ['0', '1', '2', '3', '4']
for cross_val_index in cross_val_lists:
    log.write('\n\ncross_val_index: ' + cross_val_index + '\n\n')
    if True:
        trainval_test(cross_val_index, sigma=30 * 0.1, lam=6 * 0.1)

 包含图片即标签的文本文件

    TRAIN_FILE = './Classification/NNEW_trainval_' + cross_val_index + '.txt'
    TEST_FILE = './Classification/NNEW_test_' + cross_val_index + '.txt'

生成返回图片、标签及痤疮个数的数据集集合

    normalize = transforms.Normalize(mean=[0.45815152, 0.361242, 0.29348266],
                                     std=[0.2814769, 0.226306, 0.20132513])

    dset_train = dataset_processing.DatasetProcessing(
        DATA_PATH, TRAIN_FILE, transform=transforms.Compose([
                transforms.Scale((256, 256)),
                transforms.RandomCrop(224),
                transforms.RandomHorizontalFlip(),
                transforms.ToTensor(),
                RandomRotate(rotation_range=20),
                normalize,
            ])) # return img, label, lesion

    dset_test = dataset_processing.DatasetProcessing(
        DATA_PATH, TEST_FILE, transform=transforms.Compose([
                transforms.Scale((224, 224)),
                transforms.ToTensor(),
                normalize,
            ]))

 打包batch

    train_loader = DataLoader(dset_train,
                              batch_size=BATCH_SIZE,
                              shuffle=True,
                              num_workers=NUM_WORKERS,
                              pin_memory=False) # 将训练数据集封装为一个可迭代的批处理对象

    test_loader = DataLoader(dset_test,
                             batch_size=BATCH_SIZE_TEST,
                             shuffle=False,
                             num_workers=NUM_WORKERS,
                             pin_memory=False)

120个epoch，每个epoch下，迭代所有的batch