Pass all forward op test

* Correct mul_op implementation

* Restore the origin shape after mul

* Fix mul op

* Do not touch math_function

* init

* ensure ids in lookup table op must be a column vector

* add book4 configuration in test_layers

* debug test_book4

* add test_word2vec

* follow comments

* follow comments

* ensure ids in lookup table op must be a column vector

* follow comments

* Init file

* Update

* Update

* Complete conv net of MNIST

* Adding incremnt op

* Fixing comment about step attribute

* Adding the interface for the momentum optimizer
* Adding a comment about accumulators

* Unify `set_feed_variable` to one method

* Move global scope to python, not in C++

* add test_fit_a_line

* Update

* fix persistable bug

* fix elementwise add bug

* set correct attr for bias op in fc layer

* set correct attr for bias op in fc layer

* Update

1. Add init_program to hold initializers
2. bug fix

* add test_fit_a_line

* fix persistable bug

* fix elementwise add bug

* fix type

* add gitignore

* Complete fit_a_line test

* revert code

* Clean up

* Revert "revert code"

This reverts commit eb1aa015.

* Refine

* Fix unit test

* Init

* unify layer names

* Update

* Add pool2d layer

* Test recognize_digits_conv

* Clean up

* Implement FC layer with helper

* Update LayerHelper

* Add debug string for Python ProtoBuf

and Rename `Sync` to `Flush`

* Add check of ProtoBuf initialization

* Layer wrapper for FC

* Fix unittest

* Fix CI

* Add code generator

* AttributeChecker Better error log and speicalize bool

Since lots of types can be cast to bool

* Complete mlp, fit_a_line

* Expose get global scope

* Make global scope not thread-safe

1. It is no need to make global scope thread-safe, since it will be
invoked in Python main thread.
2. Do not free the global scope when C++ exit. Let the OS free memories,
otherwise, we need to handle the destroy dependencies.

See
https://google.github.io/styleguide/cppguide.html#Static_and_Global_Variables

* Fix

* Implementation of simple conv_2d layer

* Stash

* Remove private data members in OpRegister

* Fix bugs

* Stash

* Expose FeedFetchList as VarType

* Change ProgramDesc not a global variable

* Polish code style

* Stash

* Correct implement BlockDesc destructor

* Correct implement BlockDesc destructor

* Unify program as parameter name

* Fix bugs

* Add unittest

* Fix unit test error

* Remove unused functions

* Add clone for Python Program

* Working on executor

* Stash

* Add glog as dependencies of ops

* Use VLOG to logging some information is helpful when we debug Paddle

* Expose VarDesc::persistable to Python

* Test executor

* Complete unittest

* Polish code

* Fix merge error

* Follow comment

* Polish Python Code

* remove Program.instance

* fix test_program.py

* Implement FC layer with helper

* Update LayerHelper

* Add debug string for Python ProtoBuf

and Rename `Sync` to `Flush`

* Add check of ProtoBuf initialization

* Layer wrapper for FC

* Fix unittest

* Fix CI

* Add code generator

* AttributeChecker Better error log and speicalize bool

Since lots of types can be cast to bool

* Complete mlp, fit_a_line

* Implementation of simple conv_2d layer

* Fix bugs

* Change ProgramDesc not a global variable

* Polish code style

* Stash

* Correct implement BlockDesc destructor

* Correct implement BlockDesc destructor

* Unify program as parameter name

* Fix bugs

* Add unittest

* Fix unit test error

* Remove unused functions

* Add clone for Python Program

* Compare OpDescBind directly

* Adding Proximal Gradient Descent

* Fixing review comments

* Implement FC layer with helper

* Update LayerHelper

* Add debug string for Python ProtoBuf

and Rename `Sync` to `Flush`

* Add check of ProtoBuf initialization

* Layer wrapper for FC

* Fix unittest

* Fix CI

* Add code generator

* AttributeChecker Better error log and speicalize bool

Since lots of types can be cast to bool

* Complete mlp, fit_a_line

* Implementation of simple conv_2d layer

* Fix bugs

* Correct implement BlockDesc destructor

* Fix bugs

* Fix unit test error

* Follow comments

* Change ProgramDesc not a global variable

* Polish code style

* Correct implement BlockDesc destructor

* Unify program as parameter name