我有三个数据框用于机器学习程序的训练、验证和测试。这些数据框是从读取csv文件的pandas数据框中分割出来的。以下是该文件的一个样本：

id,label904797024fe2c8ebe4c12f54baf34c62c05ec1ff,10ad0a93569e96a95ed1e777b983452e9dbd445f9,0

ID是图像文件名，不包括扩展名.tif。

之前，我的训练和验证数据在同一个数据框中，但为了避免方差，我将它们分成了两个数据框。

这是我之前的代码：

train_datagen = ImageDataGenerator(rescale = 1./255,                                  validation_split = 0.1)test_datagen = ImageDataGenerator(rescale = 1./255)train_val_path = "../input/train/"train_generator = train_datagen.flow_from_dataframe(                dataframe = df_train_val,                directory = train_val_path,                x_col = "id",                y_col = "label",                subset = "training",                target_size = (96, 96),                batch_size = 32,                class_mode="binary",                validate_filenames=False                )validation_generator = train_datagen.flow_from_dataframe(dataframe = df_train_val,                                                     directory = train_val_path,                                                        x_col = "id",                                                        y_col = "label",                                                        subset = "validation",                                                        target_size = (96, 96),                                                        batch_size = 32,                                                        class_mode="binary",                                                        validate_filenames=False                                                        )

如您在第一行所见，ImageDataGenerator设置了0.1的验证分割。如果我已经进行了这种分割，我该如何调整代码使其工作？

回答：

默认情况下，验证分割为0.0，即没有样本用于验证。

train_datagen = ImageDataGenerator(rescale = 1./255)test_datagen = ImageDataGenerator(rescale = 1./255) # 不需要train_val_path = "../input/train/"train_generator = train_datagen.flow_from_dataframe(                dataframe = df_train_val,                directory = train_val_path,                x_col = "id",                y_col = "label",                subset = None,         # 因为未指定验证分割                target_size = (96, 96),                batch_size = 32,                class_mode="binary",                validate_filenames=False                )validation_generator = train_datagen.flow_from_dataframe(dataframe = df_train_val,   # 传递测试数据框                                                     directory = train_val_path,                                                        x_col = "id",                                                        y_col = "label",                                                        subset = None,       # 因为未指定验证分割                                                        target_size = (96, 96),                                                        batch_size = 32,                                                        class_mode="binary",                                                        validate_filenames=False                                                        )