Since asking the question I have continued to find the answer. Maybe someone is interested in the answer, too.

From here: https://developers.google.com/protocol-buffers/docs/reference/cpp-generated

&gt; void set_allocated_foo(string* value): Sets the string object to the
&gt; field and frees the previous field value if it exists. If the string
&gt; pointer is not NULL, the message takes ownership of the allocated
&gt; string object and has_foo() will return true. Otherwise, if the value
&gt; is NULL, the behavior is the same as calling clear_foo(). string*
&gt;
&gt; release_foo(): Releases the ownership of the field and returns the
&gt; pointer of the string object. After calling this, caller takes the
&gt; ownership of the allocated string object, has_foo() will return false,
&gt; and foo() will return the default value.

Which means: As long as you do **not** call `release_*`, protobuf will take care of deleting the object. If you need the Object after dealing with the Protobuf Message, you need to relase it using `release_*`, which will prevent Protobuf to delete your object.

My concern is how to fit image using  `android:scaleType=&quot;fitXY&quot;` into image using **Glide**.

My `ImageView` is 


    &lt;ImageView
        android:id=&quot;@+id/img_pager&quot;
        android:layout_width=&quot;wrap_content&quot;
        android:layout_height=&quot;wrap_content&quot;
        android:layout_centerInParent=&quot;true&quot;
        android:scaleType=&quot;fitXY&quot; /&gt;

Loads image using **Glide** like this


 

    Glide.with(context).load(url).placeholder(R.drawable.default_image).into(img);


but image is not getting fit into `ImageView` it show space on image either side as shown in screen I need it `fitXY`

[![enter image description here][1]][1]


  [1]: http://i.stack.imgur.com/ariWk.png

How to fit Image into ImageView using Glide

I have a string representing the current time: `2015-11-24T19:40:00`. How do I parse this string in Javascript to get a Date represented by this string as the **LOCAL TIME**?  **Due to some restriction, I cannot use the library `moment`, but jquery is allowed. I know that someone has asked this question before, but the answer used `moment`**

For example, if I run the script in California, then this string would represent 7PM pacific time, but if I run the script in NY then this string would represent Eastern Time?

I tried the following but Chrome and Firefox give me different results:

    var str = &quot;2015-11-24T19:40:00&quot;;
    var date = new Date(str);

Chrome consumes it as **UTC** time (`Tue Nov 24 2015 11:40:00 GMT-0800 (Pacific Standard Time)`), 

but Firefox consumes it as my local **PACIFIC** time (`Tue Nov 24 2015 19:40:00 GMT-0800 (Pacific Standard Time)`)

I tried adding &quot;Z&quot; to `str`, like this `var date = new Date(str+&quot;Z&quot;);`, then both browsers give me UTC time. Is there any similar letter to `&quot;Z&quot;` which tells all browsers (at least chrome, Firefox and Safari) to parse the string as local time zone?

Javascript: parse a string to Date as LOCAL time zone

I have this small protobuf code (simplified, only the necessary is contained):

    message ParamsMessage {
    	required int32 temperature = 1;
    }
    
    message MasterMessage {
    	enum Type { GETPARAMS = 1; SENDPARAMS = 2;}
    	required Type type = 1;
    
    	optional ParamsMessage paramsMessage = 2;
    
    }

I now create a MasterMessage in the following way:

    ParamsMessage * params = new ParamsMessage();
    params-&gt;set_temperature(22);
    MasterMessage master;
    master.set_type(MasterMessage::SENDPARAMS);
    master.set_allocated_paramsmessage(params);

The question is: Do I have to (after dealing with the message) delete the `params` Message, or will protobuf delete it for me? I cannot find anything in the docs.

Protobuf: Will set_allocated_* delete the allocated object?

I have this small protobuf code (simplified, only the necessary is contained):
<pre><code class="hljs language-ini">message ParamsMessage {
	required int32 temperature = 1;
}

message MasterMessage {
	enum Type { GETPARAMS = 1; SENDPARAMS = 2;}
	required Type type = 1;

	optional ParamsMessage paramsMessage = 2;

}
</code></pre>
I now create a MasterMessage in the following way:
<pre><code class="hljs language-ini">ParamsMessage * params = new ParamsMessage();
params->set_temperature(22);
MasterMessage master;
master.set_type(MasterMessage::SENDPARAMS);
master.set_allocated_paramsmessage(params);
</code></pre>
The question is: Do I have to (after dealing with the message) delete the <code>params</code> Message, or will protobuf delete it for me? I cannot find anything in the docs.

At the very end of Scott Schurr&#39;s talk [&quot;Introducing `constexpr`&quot; at CppCon][1], he asks &quot;Is there a way to poison a function&quot;? He then explains that this can be done (albeit in a non-standard way) by:

1. Putting a `throw` in a `constexpr` function
2. Declaring an unresolved `extern const char*`
3. Referencing the unresolved `extern` in the `throw`

I sense that I&#39;m a bit out of my depth here, but I&#39;m curious:

- What does it mean to &quot;poison a function&quot;?
- What is the significance/usefulness of the technique he outlines?

  [1]: https://youtu.be/fZjYCQ8dzTc?t=3679 &quot;Cppcon&quot;


What does it mean to &quot;poison a function&quot; in C++?

# The naive, optimistic and oh.. so wrong view of the c++11 uniform initialization syntax

I thought that since C++11 user-defined type objects should be constructed with the new `{...}` syntax instead of the old `(...)` syntax (except for constructor overloaded for `std::initializer_list` and similar parameters (e.g. `std::vector`: size ctor vs 1 elem init_list ctor)).

The benefits are: no narrow implicit conversions, no problem with the most vexing parse, consistency(?). I saw no problem as I thought they are the same (except the example given).

But they are not.

# A tale of pure madness

The `{}` calls the default constructor.

... Except when:

+ the default constructor is deleted and
+ there are no other constructors defined.

Then it looks like it it rather value initializes the object?... Even if the object has deleted default constructor, the `{}` can create an object. Doesn&#39;t this beat the whole purpose of a deleted constructor?

...Except when:

+ the object has a deleted default constructor and
+ other constructor(s) defined.

Then it fails with `call to deleted constructor`.

...Except when:

+ the object has a deleted constructor and
+ no other constructor defined and
+ at least a non-static data member.

Then it fails with missing field initializers.

But then you can use `{value}` to construct the object.

*Ok maybe this is the same as the first exception (value init the object)*

...Except when:

+ the class has a deleted constructor
+ and at least one data members in-class default initialized.

Then nor `{}` nor `{value}` can create an object.

I am sure I missed a few. The irony is that it is called *uniform* initialization syntax. I say again: ***UNIFORM*** initialization syntax.

What is this madness?

## Scenario A

### Deleted default constructor:

    struct foo {
      foo() = delete;
    };

    // All bellow OK (no errors, no warnings)
    foo f = foo{};
    foo f = {};
    foo f{}; // will use only this from now on.
    
## Scenario B
### Deleted default constructor, other constructors deleted

    struct foo {
      foo() = delete;
      foo(int) = delete;
    };
    
    foo f{}; // OK

## Scenario C
### Deleted default constructor, other constructors defined

    struct foo {
      foo() = delete;
      foo(int) {};
    };
    
    foo f{}; // error call to deleted constructor

## Scenario D
### Deleted default constructor, no other constructors defined, data member

    struct foo {
      int a;
      foo() = delete;
    };
    
    foo f{}; // error use of deleted function foo::foo()
    foo f{3}; // OK

## Scenario E
### Deleted default constructor, deleted T constructor, T data member

    struct foo {
      int a;
      foo() = delete;
      foo(int) = delete;
    };
    
    foo f{}; // ERROR: missing initializer
    foo f{3}; // OK

## Scenario F
### Deleted default constructor, in-class data member initializers

    struct foo {
      int a = 3;
      foo() = delete;
    };
    
    /* Fa */ foo f{}; // ERROR: use of deleted function `foo::foo()`
    /* Fb */ foo f{3}; // ERROR: no matching function to call `foo::foo(init list)`

# 

Deleted default constructor. Objects can still be created... sometimes

I am currently studying how to write efficient C++ code, and on the matter of function calls, a question comes to mind. Comparing this pseudocode function:

    not-void function-name () {
        do-something
        return value;
    }
    int main () {
        ...
        arg = function-name();
        ...
    }

with this otherwise-identical pseudocode function:

    void function-name (not-void&amp; arg) {
        do-something
        arg = value;
    }
    int main () {
        ...
        function-name(arg);
        ...
    }

Which version is more efficient, and in what respect (time, memory etc.)? If it depends, then when would the first be more efficient and when would the more efficient be the second?

**Edit**: For context, this question is limited to hardware platform-independent differences, and for the most part software too. Are there any machine-independent performance difference?

**Edit**: I don&#39;t see how this is a duplicate. The other question is comparing passing by reference (prev. code) to passing by value (below):

    not-void function-name (not-void arg)
Which is not the same thing as my question. My focus is not on which is the better way to pass in an argument to a function. My focus is on which is the better way to pass **out** a result to a variable from the outside scope.

Which is more efficient: Return a value vs. Pass by reference?

I have seen the following macro definitions in a coding book.

    #define TRUE  &#39;/&#39;/&#39;/&#39;
    #define FALSE &#39;-&#39;-&#39;-&#39;

There was no explanation there.

Please explain to me how these will work as `TRUE` and `FALSE`.

Strange definitions of TRUE and FALSE macros

Why can&#39;t you pass an object by reference when creating a ```std::thread``` ?

For example the following snippit gives a compile error:

    #include &lt;iostream&gt;
    #include &lt;thread&gt;
    
    using namespace std;
    
    static void SimpleThread(int&amp; a)  // compile error
    //static void SimpleThread(int a)     // OK
    {
        cout &lt;&lt; __PRETTY_FUNCTION__ &lt;&lt; &quot;:&quot; &lt;&lt; a &lt;&lt; endl;
    }
    
    int main()
    {
        int a = 6;
    
        auto thread1 = std::thread(SimpleThread, a);
        
        thread1.join();
        return 0;
    }

Error:

    In file included from /usr/include/c++/4.8/thread:39:0,
                     from ./std_thread_refs.cpp:5:
    /usr/include/c++/4.8/functional: In instantiation of ‘struct std::_Bind_simple&lt;void (*(int))(int&amp;)&gt;’:
    /usr/include/c++/4.8/thread:137:47:   required from ‘std::thread::thread(_Callable&amp;&amp;, _Args&amp;&amp; ...) [with _Callable = void (&amp;)(int&amp;); _Args = {int&amp;}]’
    ./std_thread_refs.cpp:19:47:   required from here
    /usr/include/c++/4.8/functional:1697:61: error: no type named ‘type’ in ‘class std::result_of&lt;void (*(int))(int&amp;)&gt;’
           typedef typename result_of&lt;_Callable(_Args...)&gt;::type result_type;
                                                                 ^
    /usr/include/c++/4.8/functional:1727:9: error: no type named ‘type’ in ‘class std::result_of&lt;void (*(int))(int&amp;)&gt;’
             _M_invoke(_Index_tuple&lt;_Indices...&gt;)
             ^

I&#39;ve changed to passing a pointer, but is there a better work around?

Passing object by reference to std::thread in C++11

I&#39;m looking for a command-line utility that will, at a minimum, render binary protobuf data in human-readable form.  Filtering and selection options (along the lines of `cut` for text) would be nice, but the primary object is to make the data visible for debugging purposes.

If there is no definitive tool for the job, links to relevant packages are fine.

Is there a definitive *nix command line tool for inspecting protocol buffers?

We want to build a Javascript/HTML gui for our gRPC-microservices. Since gRPC is not supported on the browser side, we thought of using web-sockets to connect to a node.js server, which calls the target service via grpc. 
We struggle to find an elegant solution to do this. Especially, since we use gRPC streams to push events between our micro-services.
It seems that we need a second RPC system, just to communicate between the front end and the node.js server. This seems to be a lot of overhead and additional code that must be maintained.

Does anyone have experience doing something like this or has an idea how this could be solved?

How to bring a gRPC defined API to the web browser

I want to attach a UUID to a field in my protobuf User message example.

    message User {
      // field containing id as UUID type
      required string email;
      optional string name;
    }

I know that protobuf messages do not yet support the UUID type. I&#39;ve read that the best approach is to have a UUID message type.

So I&#39;m guessing my User message would import my UUID message proto definition and use it as a field type like so:

    import &quot;myproject/UUID.proto&quot;;
    
    message User {
      required UUID id;
      required string email;
      optional string name;
    }

My question is, how will the UUID message look like, and how will I encode/decode it? I&#39;m aiming for Java/Scala and C# compatibility.

How do I represent a UUID in a protobuf message?

I am following [this guide](https://developers.google.com/protocol-buffers/docs/pythontutorial) and using the exact sample of addressbook.proto.

Post the content of compiler generated addressbook_pb2.py file as well.
When I run the following simple program, there is error saying, cannot find google.protobuf, any ideas how to resolve this issue? Thanks.

BTW, using Python 2.7 on Mac OSX.

    from addressbook_pb2 import Person
    
    p = Person()
    p.email = &quot;abc&quot;
    
    print p.email


**Here is the automated generated file addressbook_pb2.py,**

    # Generated by the protocol buffer compiler.  DO NOT EDIT!
    # source: addressbook.proto
    
    import sys
    _b=sys.version_info[0]&lt;3 and (lambda x:x) or (lambda x:x.encode(&#39;latin1&#39;))
    from google.protobuf import descriptor as _descriptor
    from google.protobuf import message as _message
    from google.protobuf import reflection as _reflection
    from google.protobuf import symbol_database as _symbol_database
    from google.protobuf import descriptor_pb2
    # @@protoc_insertion_point(imports)
    
    _sym_db = _symbol_database.Default()
    
    
    
    
    DESCRIPTOR = _descriptor.FileDescriptor(
      name=&#39;addressbook.proto&#39;,
      package=&#39;tutorial&#39;,
      syntax=&#39;proto2&#39;,
      serialized_pb=_b(&#39;\n\x11\x61\x64\x64ressbook.proto\x12\x08tutorial\&quot;\xda\x01\n\x06Person\x12\x0c\n\x04name\x18\x01 \x02(\t\x12\n\n\x02id\x18\x02 \x02(\x05\x12\r\n\x05\x65mail\x18\x03 \x01(\t\x12+\n\x05phone\x18\x04 \x03(\x0b\x32\x1c.tutorial.Person.PhoneNumber\x1aM\n\x0bPhoneNumber\x12\x0e\n\x06number\x18\x01 \x02(\t\x12.\n\x04type\x18\x02 \x01(\x0e\x32\x1a.tutorial.Person.PhoneType:\x04HOME\&quot;+\n\tPhoneType\x12\n\n\x06MOBILE\x10\x00\x12\x08\n\x04HOME\x10\x01\x12\x08\n\x04WORK\x10\x02\&quot;/\n\x0b\x41\x64\x64ressBook\x12 \n\x06person\x18\x01 \x03(\x0b\x32\x10.tutorial.Person&#39;)
    )
    _sym_db.RegisterFileDescriptor(DESCRIPTOR)
    
    
    
    _PERSON_PHONETYPE = _descriptor.EnumDescriptor(
      name=&#39;PhoneType&#39;,
      full_name=&#39;tutorial.Person.PhoneType&#39;,
      filename=None,
      file=DESCRIPTOR,
      values=[
        _descriptor.EnumValueDescriptor(
          name=&#39;MOBILE&#39;, index=0, number=0,
          options=None,
          type=None),
        _descriptor.EnumValueDescriptor(
          name=&#39;HOME&#39;, index=1, number=1,
          options=None,
          type=None),
        _descriptor.EnumValueDescriptor(
          name=&#39;WORK&#39;, index=2, number=2,
          options=None,
          type=None),
      ],
      containing_type=None,
      options=None,
      serialized_start=207,
      serialized_end=250,
    )
    _sym_db.RegisterEnumDescriptor(_PERSON_PHONETYPE)
    
    
    _PERSON_PHONENUMBER = _descriptor.Descriptor(
      name=&#39;PhoneNumber&#39;,
      full_name=&#39;tutorial.Person.PhoneNumber&#39;,
      filename=None,
      file=DESCRIPTOR,
      containing_type=None,
      fields=[
        _descriptor.FieldDescriptor(
          name=&#39;number&#39;, full_name=&#39;tutorial.Person.PhoneNumber.number&#39;, index=0,
          number=1, type=9, cpp_type=9, label=2,
          has_default_value=False, default_value=_b(&quot;&quot;).decode(&#39;utf-8&#39;),
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
        _descriptor.FieldDescriptor(
          name=&#39;type&#39;, full_name=&#39;tutorial.Person.PhoneNumber.type&#39;, index=1,
          number=2, type=14, cpp_type=8, label=1,
          has_default_value=True, default_value=1,
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
      ],
      extensions=[
      ],
      nested_types=[],
      enum_types=[
      ],
      options=None,
      is_extendable=False,
      syntax=&#39;proto2&#39;,
      extension_ranges=[],
      oneofs=[
      ],
      serialized_start=128,
      serialized_end=205,
    )
    
    _PERSON = _descriptor.Descriptor(
      name=&#39;Person&#39;,
      full_name=&#39;tutorial.Person&#39;,
      filename=None,
      file=DESCRIPTOR,
      containing_type=None,
      fields=[
        _descriptor.FieldDescriptor(
          name=&#39;name&#39;, full_name=&#39;tutorial.Person.name&#39;, index=0,
          number=1, type=9, cpp_type=9, label=2,
          has_default_value=False, default_value=_b(&quot;&quot;).decode(&#39;utf-8&#39;),
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
        _descriptor.FieldDescriptor(
          name=&#39;id&#39;, full_name=&#39;tutorial.Person.id&#39;, index=1,
          number=2, type=5, cpp_type=1, label=2,
          has_default_value=False, default_value=0,
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
        _descriptor.FieldDescriptor(
          name=&#39;email&#39;, full_name=&#39;tutorial.Person.email&#39;, index=2,
          number=3, type=9, cpp_type=9, label=1,
          has_default_value=False, default_value=_b(&quot;&quot;).decode(&#39;utf-8&#39;),
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
        _descriptor.FieldDescriptor(
          name=&#39;phone&#39;, full_name=&#39;tutorial.Person.phone&#39;, index=3,
          number=4, type=11, cpp_type=10, label=3,
          has_default_value=False, default_value=[],
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
      ],
      extensions=[
      ],
      nested_types=[_PERSON_PHONENUMBER, ],
      enum_types=[
        _PERSON_PHONETYPE,
      ],
      options=None,
      is_extendable=False,
      syntax=&#39;proto2&#39;,
      extension_ranges=[],
      oneofs=[
      ],
      serialized_start=32,
      serialized_end=250,
    )
    
    
    _ADDRESSBOOK = _descriptor.Descriptor(
      name=&#39;AddressBook&#39;,
      full_name=&#39;tutorial.AddressBook&#39;,
      filename=None,
      file=DESCRIPTOR,
      containing_type=None,
      fields=[
        _descriptor.FieldDescriptor(
          name=&#39;person&#39;, full_name=&#39;tutorial.AddressBook.person&#39;, index=0,
          number=1, type=11, cpp_type=10, label=3,
          has_default_value=False, default_value=[],
          message_type=None, enum_type=None, containing_type=None,
          is_extension=False, extension_scope=None,
          options=None),
      ],
      extensions=[
      ],
      nested_types=[],
      enum_types=[
      ],
      options=None,
      is_extendable=False,
      syntax=&#39;proto2&#39;,
      extension_ranges=[],
      oneofs=[
      ],
      serialized_start=252,
      serialized_end=299,
    )
    
    _PERSON_PHONENUMBER.fields_by_name[&#39;type&#39;].enum_type = _PERSON_PHONETYPE
    _PERSON_PHONENUMBER.containing_type = _PERSON
    _PERSON.fields_by_name[&#39;phone&#39;].message_type = _PERSON_PHONENUMBER
    _PERSON_PHONETYPE.containing_type = _PERSON
    _ADDRESSBOOK.fields_by_name[&#39;person&#39;].message_type = _PERSON
    DESCRIPTOR.message_types_by_name[&#39;Person&#39;] = _PERSON
    DESCRIPTOR.message_types_by_name[&#39;AddressBook&#39;] = _ADDRESSBOOK
    
    Person = _reflection.GeneratedProtocolMessageType(&#39;Person&#39;, (_message.Message,), dict(
    
      PhoneNumber = _reflection.GeneratedProtocolMessageType(&#39;PhoneNumber&#39;, (_message.Message,), dict(
        DESCRIPTOR = _PERSON_PHONENUMBER,
        __module__ = &#39;addressbook_pb2&#39;
        # @@protoc_insertion_point(class_scope:tutorial.Person.PhoneNumber)
        ))
      ,
      DESCRIPTOR = _PERSON,
      __module__ = &#39;addressbook_pb2&#39;
      # @@protoc_insertion_point(class_scope:tutorial.Person)
      ))
    _sym_db.RegisterMessage(Person)
    _sym_db.RegisterMessage(Person.PhoneNumber)
    
    AddressBook = _reflection.GeneratedProtocolMessageType(&#39;AddressBook&#39;, (_message.Message,), dict(
      DESCRIPTOR = _ADDRESSBOOK,
      __module__ = &#39;addressbook_pb2&#39;
      # @@protoc_insertion_point(class_scope:tutorial.AddressBook)
      ))
    _sym_db.RegisterMessage(AddressBook)
    
    
    # @@protoc_insertion_point(module_scope)

**Edit 1**

Tried `pip install protobuf`, met with the following error,

    Requirement already satisfied (use --upgrade to upgrade): protobuf in /Users/foo/miniconda2/lib/python2.7/site-packages
    Requirement already satisfied (use --upgrade to upgrade): six&gt;=1.9 in /Users/foo/miniconda2/lib/python2.7/site-packages/six-1.10.0-py2.7.egg (from protobuf)
    Requirement already satisfied (use --upgrade to upgrade): setuptools in /Users/foo/miniconda2/lib/python2.7/site-packages (from protobuf)

Here is the output of python version,

    python -V
    Python 2.7.11 :: Continuum Analytics, Inc.

**Edit 2**

Post exact error message,

    Traceback (most recent call last):
      File &quot;/Applications/PyCharm CE.app/Contents/helpers/pydev/pydevd.py&quot;, line 1531, in &lt;module&gt;
        globals = debugger.run(setup[&#39;file&#39;], None, None, is_module)
      File &quot;/Applications/PyCharm CE.app/Contents/helpers/pydev/pydevd.py&quot;, line 938, in run
        pydev_imports.execfile(file, globals, locals)  # execute the script
      File &quot;/Users/foo/personal/featureExtraction/protobuf_test.py&quot;, line 1, in &lt;module&gt;
        from addressbook_pb2 import Person
      File &quot;/Applications/PyCharm CE.app/Contents/helpers/pydev/_pydev_bundle/pydev_monkey_qt.py&quot;, line 71, in patched_import
        return original_import(name, *args, **kwargs)
      File &quot;/Users/foo/personal/featureExtraction/addressbook_pb2.py&quot;, line 6, in &lt;module&gt;
        from google.protobuf import descriptor as _descriptor
      File &quot;/Applications/PyCharm CE.app/Contents/helpers/pydev/_pydev_bundle/pydev_monkey_qt.py&quot;, line 71, in patched_import
        return original_import(name, *args, **kwargs)
    ImportError: No module named google.protobuf

**Edit 3**

error message when `import google`,

[![enter image description here][1]][1]

**Edit 4**

Output of `which pip`,

    which pip
    /Users/foo/miniconda2/bin/pip

Output of `sys.executable`,

/Users/foo/anaconda/bin/python

**Edit 5**

    foo-mn1:featureExtraction foo$ sudo /Users/foo/miniconda2/bin/pip install protobuf
    Password:
    The directory &#39;/Users/foo/Library/Caches/pip/http&#39; or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo&#39;s -H flag.
    The directory &#39;/Users/foo/Library/Caches/pip&#39; or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo&#39;s -H flag.
    Requirement already satisfied (use --upgrade to upgrade): protobuf in /Users/foo/miniconda2/lib/python2.7/site-packages
    Requirement already satisfied (use --upgrade to upgrade): six&gt;=1.9 in /Users/foo/miniconda2/lib/python2.7/site-packages/six-1.10.0-py2.7.egg (from protobuf)
    Requirement already satisfied (use --upgrade to upgrade): setuptools in /Users/foo/miniconda2/lib/python2.7/site-packages (from protobuf)
    foo-mn1:featureExtraction foo$ sudo /Users/foo/miniconda2/bin/pip install google
    The directory &#39;/Users/foo/Library/Caches/pip/http&#39; or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo&#39;s -H flag.
    The directory &#39;/Users/foo/Library/Caches/pip&#39; or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo&#39;s -H flag.
    Requirement already satisfied (use --upgrade to upgrade): google in /Users/foo/miniconda2/lib/python2.7/site-packages
    Requirement already satisfied (use --upgrade to upgrade): beautifulsoup4 in /Users/foo/miniconda2/lib/python2.7/site-packages (from google)

  [1]: http://i.stack.imgur.com/v2gS3.png

ImportError: No module named google.protobuf

From what I&#39;ve gathered so far, there are several different ways of dumping a TensorFlow graph into a file and then loading it into another program, but I haven&#39;t been able to find clear examples/information on how they work. What I already know is this:

1. Save the model&#39;s variables into a checkpoint file (.ckpt) using a `tf.train.Saver()` and restore them later ([source][1])
2. Save a model into a .pb file and load it back in using `tf.train.write_graph()` and `tf.import_graph_def()` ([source][2])
3. Load in a model from a .pb file, retrain it, and dump it into a new .pb file using Bazel ([source][3])
4. Freeze the graph to save the graph and weights together ([source][4])
5. Use `as_graph_def()` to save the model, and for weights/variables, map them into constants ([source][5])

However, I haven&#39;t been able to clear up several questions regarding these different methods:

1. Regarding checkpoint files, do they only save the trained weights of a model? Could checkpoint files be loaded into a new program, and be used to run the model, or do they simply serve as ways to save the weights in a model at a certain time/stage?
2. Regarding `tf.train.write_graph()`, are the weights/variables saved as well?
3. Regarding Bazel, can it only save into/load from .pb files for retraining? Is there a simple Bazel command just to dump a graph into a .pb?
4. Regarding freezing, can a frozen graph be loaded in using `tf.import_graph_def()`?
5. The Android demo for TensorFlow loads in Google&#39;s Inception model from a .pb file. If I wanted to substitute my own .pb file, how would I go about doing that? Would I need to change any native code/methods?
6. In general, what exactly is the difference between all these methods? Or more broadly, what is the difference between `as_graph_def()`/.ckpt/.pb?

In short, what I&#39;m looking for is a method to save both a graph (as in, the various operations and such) and its weights/variables into a file, which can then be used to load the graph and weights into another program, for use (not necessarily continuing/retraining).

Documentation about this topic isn&#39;t very straightforward, so any answers/information would be greatly appreciated.


  [1]: https://www.tensorflow.org/versions/r0.10/how_tos/variables/index.html
  [2]: https://github.com/tensorflow/tensorflow/issues/616
  [3]: https://petewarden.com/2016/02/28/tensorflow-for-poets/
  [4]: https://www.tensorflow.org/versions/r0.9/how_tos/tool_developers/index.html#freezing
  [5]: https://stackoverflow.com/questions/34343259/is-there-an-example-on-how-to-generate-protobuf-files-holding-trained-tensorflow

Content Type	Original Author	Original Content on Stackoverflow
Question	Nidhoegger	View Question on Stackoverflow
Solution 1 - C++	Nidhoegger	View Answer on Stackoverflow

Protobuf: Will set_allocated_* delete the allocated object?

C++ Problem Overview

C++ Solutions

Solution 1 - C++

Javascript: parse a string to Date as LOCAL time zone

How to fit Image into ImageView using Glide

Attributions