小编081*_*张庆昊的帖子

如果我运行此代码两次:

tf.flags.DEFINE_integer("batch_size", "2", "batch size for training")

我会收到这个错误:

DuplicateFlagError: The flag 'batch_size' is defined twice. First from D:/Python/workspace/FCN_dataset/FCN.tensorflow-master/FCN.py, Second from D:/Python/workspace/FCN_dataset/FCN.tensorflow-master/FCN.py.  Description from first occurrence: batch size for training

我知道这是因为我重新定义了flag的默认值.那么如何清除标志或允许重新定义标志的默认值呢？

像这样：

typeArray type[2] = {uint8_t, uint16_t};
for (int i=0; i < 2; i++)
{
    myFunction<type[i]>();
}

我认为这是不可能实现的，因为它违反了 C++ 中的一些规则。但我不确定，我也不熟悉 C++ 的新特性（11 14 17 等）。所以有人可以告诉我答案吗？

更加具体。

\\ Such as I have a template functon "PrintSecondTypeLen"
template<typename Type>PrintSecondTypeLen(unsigned int *a)
{
    Type *b;
    b = (Type *) a;
    std::cout << *(b+1) << std::endl;
}
\\ Then suppose that I have a typeArray.
typeArray type[2] = {uint8_t, uint16_t}; \\ It's wrong, just for example.

\\ Then I use it:
unsigned int a[2] = …

我尝试将 python 脚本添加到我的包中，参考这两个教程。

处理 setup.py

安装 Python 脚本和模块

所以我在 root 中添加了 setup.py ，test\src\test_pkg在 path 中更改了 CMakeLists.txt test\src。（我的包根路径是test\，我的包路径是test\src\test_pkg，我的python脚本路径是test\src\test_pkg\scripts）

这是 setup.py。

#!/usr/bin/env python
# -*- coding: utf-8 -*-


from distutils.core import setup
from catkin_pkg.python_setup import generate_distutils_setup


setup_args = generate_distutils_setup(
    packages=['test_pkg'],
    scripts=['/scripts'],
    package_dir={'': 'src'}
)

setup(**setup_args)

这是 CMakeLists.txt

cmake_minimum_required(VERSION 2.8.3)
project(test_pkg)

find_package(catkin REQUIRED COMPONENTS
  roscpp
  rospy
  std_msgs
  sensor_msgs
  message_generation
)

catkin_python_setup()

catkin_install_python(PROGRAMS scripts/talker
                      DESTINATION ${CATKIN_PACKAGE_BIN_DESTINATION})

add_message_files(
  FILES
  Num.msg
)

add_service_files(
  FILES
  AddTwoInts.srv
)

generate_messages( …