* Add documentation of cross-compiling for iOS.

* Correst the typo in documentation of cross-compiling for raspberry pi.

* Set ANDROID_API to 21 when it is specified < 21 for arm64-v8a in build_android.sh.

* Check the input and print the usage in MergeModel.cpp.

Add backward implementation for LSTM operator. 

Allocate aligned memory by posix_memalign.

* "replace enum with string"

* "fix layers"

Set gradient's data type based on its forward variable

* Init commit

* Make executor use ProgramDescBind

* Change Attribute from BlockDesc to BlockDescBind

* Since we will get the program desc in RNN, just BlockDesc is not
  enough.

* add shareLod

* fix sequence_conv grad infershape

1. Change some `auto` to `auto*`
2. Change `Tensor` to `LoDTensor`

* AddBiasOp does not care num_flatten_dims

* Add comments

* "modify layers.py"

* "fix pool interface"

* "add export type to layers"

* "fix based on comment"

correct the index of cluster_train_cn/en.md

Use posix_memalign to allocate aligned memory, since some SIMD instructions require the alignment of memory accesses.

add test_Expand and simply the gserver/tests/CMakeLists

* refine evaluator op types

* update

* follow comments

* update

* fix v2 mnist case

* fix v2 mnist case

* update

* update

* Complete accuray layer

* Fix error

* Fix error

* Add 'accuracy' to __all__

* update

* Fix Type error

* Fix error

* Refine unit tests

* Fix an unit test error

* Add cross_compiling_for_raspberry.md

* Update cross_compiling for raspberry pi document

* Some minor edits

* In response to comments from Kavya

* Add the _en suffix

* reimplement pow operator

* add pow_grad operator

* fix code style

* fix build error

* fix op_test bug

* revert pow operator

* add FIXME comment

* remove unused code

* fix cmake file

* fix build error

* improve unique_name, uniq id is related to prefix

* fix join

* add logs and fix a bug

* fix break buf

* modify path bugs

* fix by comments

* fix by comments

* add batch

* add float32tostring

* add pb support

* moidfy gotpaht

* compile ok

* add proto

* delete not need

* add proto

* add empty proto

* clean not need

* clean not need

* modify deps

* fix by comments and update depend

* fix compile error

* fix loop bugs

* "add sequence conv layer"

* "add sequence layer"

* add networks

* "fix based comment"

* Update layers.py

* Fix Type error

* Fix error

* Fix top_k_op GPU code data type

Update CONTRIBUTION.md to generate correctly on SPHINX