* Enhence generate_proposal_labels_op
* Fix bug in generate_proposals_op
* Refine rpn_target_assign_op.
* by Bu Xingyuan, Wang Guanzhong and Dang Qingqing

* trainer auto wait pserver port ready

* add file

* fix docstring

* add option to not wait

* update api spec

* clean

* fix test hang

import os module to use os.environ in setUp()

* add multiprocess_reader

* add multiprocess_reader to reader decorator

* support piped multi process reader

* revert v2 decorator

* add comment to multiprocess_reader

* optimize code

* use ujson to speed up json serialize/deserialize

* add assert to multiprocess_reader

* update comment of multiprocess_reader

* optimize ujson import, handle error case

* optimize import ujson

* remove ujson from requirements.txt

* add import sys to decorator.py

* add ignore index

* update api.spec

* enhance softmax_with_cross_entropy

* Implement norm_op by CUDA instead of Eigen.
* Remove the commented code.

* support test using executor without reader

* run imgnet

* update fluid benchmark

* wip

* update

* update all models

* support pyreader

* update

* clean up

* make profile batches contollable

* update API.spec

* update scripts

* clean dockerfile

* update

* clean comments

* add scope argument docstring

* use num_trainers to determine nccl init comms

* add save_inference_model for Trainer

* add comment

* update comment

* fix mac compile

* add unit test

* update API.spec

* revert cpplint-cpp-source

* Add intermediate

* fix flatten/squeeze/unsqueeze

* Considering compatibility issues, we could not fix the origin op

* follow comment

* reset the shape of XShape