add iou similarity operator

Add python API for im2sequence op

* limit variable type to lod tensor in memory optimization transpiler

* refine policy

* support while operator

* fix random seed and training data order

* refine get_cfgs method to support multi while operators

* refine codes

Disable the unit test ofdetection output op

prior box operator for ssd

Add python wrapper for GRU

Wrap NCE to python

* migrate registry server

* use 0.11.0 tag

Fix bug and unit test in bipartite_match_op.

fix SyntaxWarning in wmt16.py

* refine conv2d_transpose layer doc

* fix conv2d_transpose doc

* fix doc

Register reduce_op_kernel with multiple data types

Add bipartite matching operator and unit testing.

fix rendering error of transpose operator and add wrapper.

Add im2sequence op.

* Updating the cluster trainign doc

* Fixed few comments of transpiler

* Adding few explanations