解决方法是确保在两个函数中使用相同的batch size和shuffle参数。代码示例如下：

train_generator = train_datagen.flow_from_directory(
        'train_dir',
        target_size=(224, 224),
        batch_size=32,
        class_mode='binary')

validation_generator = test_datagen.flow_from_directory(
        'validation_dir',
        target_size=(224, 224),
        batch_size=32,
        class_mode='binary')

model.fit_generator(
        train_generator,
        epochs=10,
        validation_data=validation_generator,
        shuffle=True)

scores = model.evaluate_generator(
        validation_generator,
        steps=validation_generator.samples // 32)

print('Validation accuracy:', scores[1])