macos安装iTouch踩过的雷

啊，这个故事还是得从Torch深度学习的那个教程开始说起...

因为在参考的其中一篇论文里面看到论文作者使用的是torch7这个框架来跑神经网络，所以本菜鸡也就屁颠屁颠地跑过去，跟人家大大用一样的环境。装好环境就打算按官网教程跑一遍，本来一切都还挺顺利的，在跑到教程的显示图片那段代码中就开始出问题了。就是下面这段

itorch.image(trainset.data[100]) -- display the 100-th image in dataset
print(classes[trainset.label[100]])

本人是在sublime text上面跑的，现在忘记报什么错了，反正就是看到网上有人说这个要在notebook上跑，然后就跑去装notebook。macos自带的是python2.7，因此安装命令是

python -m pip install jupyter --user

安装成功之后运行jupyter notebook就报错:UnicodeDecodeError: 'ascii' codec can't decode byte 0xe5 in position 4: ordinal not in range(128)

详情如下：

Traceback (most recent call last):
  File "/Users/haroroda/Library/Python/2.7/bin/ipython", line 11, in <module>
    sys.exit(start_ipython())
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/IPython/__init__.py", line 119, in start_ipython
    return launch_new_instance(argv=argv, **kwargs)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 657, in launch_instance
    app.initialize(argv)
  File "<decorator-gen-110>", line 2, in initialize
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 87, in catch_config_error
    return method(app, *args, **kwargs)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/IPython/terminal/ipapp.py", line 307, in initialize
    super(TerminalIPythonApp, self).initialize(argv)
  File "<decorator-gen-7>", line 2, in initialize
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 87, in catch_config_error
    return method(app, *args, **kwargs)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/IPython/core/application.py", line 455, in initialize
    self.parse_command_line(argv)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/IPython/terminal/ipapp.py", line 302, in parse_command_line
    return super(TerminalIPythonApp, self).parse_command_line(argv)
  File "<decorator-gen-4>", line 2, in parse_command_line
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 87, in catch_config_error
    return method(app, *args, **kwargs)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 514, in parse_command_line
    return self.initialize_subcommand(subc, subargv)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/IPython/core/application.py", line 244, in initialize_subcommand
    return super(BaseIPythonApplication, self).initialize_subcommand(subc, argv)
  File "<decorator-gen-3>", line 2, in initialize_subcommand
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 87, in catch_config_error
    return method(app, *args, **kwargs)
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 452, in initialize_subcommand
    self.subapp.initialize(argv)
  File "<decorator-gen-114>", line 2, in initialize
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 89, in catch_config_error
    app.print_help()
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 385, in print_help
    self.print_subcommands()
  File "/Users/haroroda/Library/Python/2.7/lib/python/site-packages/traitlets/config/application.py", line 377, in print_subcommands
    print(os.linesep.join(lines))
UnicodeDecodeError: 'ascii' codec can't decode byte 0xe5 in position 4: ordinal not in range(128)

重头戏来了，关于这个报错我搜了一圈，好几个人都说这是python2.7带的bug，换成python3就好了。大佬都这么说了还能怎么办，只能装python3呗。一顿折腾把python3装好了，默认的python版本也换成3.7了。这时候运行jupyter notebook没报错，成功运行了。心里正美滋滋呢，接着跑去装iTorch。按官网的安装成功后运行itorch，报错

Unsupported ipython version. Only major versions 2.xx, 3.xx, 4.xx or 5.xx (Jupyter) are supported

一看ipython的版本是7.xx。用python3装的jupyter自带ipython7.xx，用python2装的jupyter带的是ipython5.xx。（本人试过单独下载ipython5.3.0的版本，忘记是哪步出错了反正也是不行）

到这里各位都看明白了把，要成功运行jupyter就不能用python2，只能用python3；但要跑通iTorch就不能使用python3，只支持python2，感觉进入了死循环了。正在本菜鸡愁眉不展的时候看到一个大大说，在运行的时候指定语言就好了。原文地址：https://www.jianshu.com/p/516eb2a57ee6

LANG=zn jupyter notebook

这样就可以在python2的环境下顺利运行jupyter，同理，运行itorch notebook命令如下：

LANG=zn itorch notebook