* Load/save the params from the params_path

* Switch to use load_persistables and save_persistables

* Instaed of setup the executor to run program and scope. Pass the program to the load_persistables

* Added auto transform to beam_search_decode_op

* Added some comment

* Added unittest for beam_search_decode_op on GPU

* add test float16 inference accuracy example

* complete the test

* clean code

* add argument parse and refine tests

* add shell script

* add float16 benchmark code

* refine code

* prepare for contrib/float16

* put things in contrib float16 folder

* update benchmark result

* further update benchmark report

* add float16 inference report

* update report

Works with 1 trainer 1 pserver. 2 trainer 1 pserver will stuck at the
end of first step, still investigating.

The user only need to set envrionment variables to enable distributed
training.

run pserver:

PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=2 PADDLE_CURRENT_IP=127.0.0.1 python no_test_word2vec_new_api.py

run trainer:

PADDLE_TRAINING_ROLE=TRAINER PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=2 PADDLE_TRAINER_ID=0 python no_test_word2vec_new_api.py

* Add understand apiu with stacked lstm for new API

* Complete exam

* Initial implementation of forward pass for MKLDNN batch norm

* Added attributes for MKLDNN batch norm

* MKLDNN batch norm forward pass passes unittest. Started working on backward

* Backward pass for MKLDNN batch norm added

* MKLDNN batch norm: scoring added to forward pass

* MKLDNN batch norm: bias as input added; handling AnyLayout when kernel is looked up

* MKLDNN batch norm: python unit tests added; mkldnn tests removed

* MKLDNN batch norm: changes required by cpplint

* MKLDNN batch norm: refactoring the operator

* MKLDNN batch norm: saved variance inversed in backward pass for correct execution of MKLDNN unit tests

* MKLDNN batch norm: refctoring, function for static/const cast to void* added

* MKLDNN batch norm: remove AnyLayout from batch norm

*  MKLDNN batch norm: only NCHW format is supported. Unittests refactored

* MKDNN batch norm: use_mkldnn added to attributes

* MKLDNN batch norm: AnyLayout removed from unittest

* MKLDNN batch norm: added CUDNN defines to batch norm

* MKLDNN batch norm: undefined data_format variable corrected

* MKLDNN batch norm: use_cudnn added, use of setUp method for configuring attributes

* MKLDNN batch norm: added use_cudnn attribute to batch norm operator

* MKLDNN batch norm: correcting batch norm unit tests for MKLDNN

* MKLDNN batch norm: MKLDNN tests moved to another file; reverting changes for saved variance not being inverted

* Change default layout to NCHW

* MKLDNN batch norm: init_kernel_type method added to unit tests

* MKLDNN batch norm: style changes

* MKLDNN batch norm: unit tests refactored

* MKLDNN batch norm: added use_mkldnn attribute to batch norm python interface

- The trainer and inferencer will load params from disk if param_path
  argument is not None in their constructor.

- Remove params.py, we will expose core.Scope to the user if needed
  (e.g., for GAN). Currently we will not expose it, unless we clearly
  know doing so can support GAN.

- Add `save_params` to Trainer (a TODO item).

- rename "network" to "program"

* "re-commit "

* "picked up"

* "fix ci"

* "fix pdb hang up issue in cuda 9"