* fix models because the gradient clip strategy has been upgraded,test=develop

* fix models because the gradient clip strategy has been upgraded,test=develop

* add cpu device count

* import multiprocessing

test=develop

* fix ptb_lm for support cpu, test=develop

* fix bug for arg_parse, test=develop

* update readme.me, test=develop

* add readme.md for ptb_lm, test=develop

* fix ptb_dy.py to eval valid_data, test=develop

* fix ptb_dy.py to eval valid_data, test=develop

* Optimizer init with parameters, test=develop

* fix bug for eval, test=develop

* update ptb model to use auto generated op functions, test=develop

* refine model code, test=develop

* refine model code, test=develop

* update ptb model, test=develop

test=develop

* update ptb model to use auto generated op functions, test=develop

* refine model code, test=develop

* refine model code, test=develop

* Use new save/load api and change fluid.layers.data to fluid.data (#3547)

* Use new save/load api and change fluid.layers.data to fluid.data

* add check for paddle version

* Use new save/load api in ptb_lm (#3546)

* Use new save/load api in ptb_lm

* add check for paddle version

* unify reader to dataloader (#3488)

unify reader to dataloader

* Use new save/load api and change fluid.layers.data to fluid.data (#3547)

* Use new save/load api and change fluid.layers.data to fluid.data

* add check for paddle version

* fix ptb reader and remove useless language model; test=develop

* add ptb reader; test=develop

* fix ptb author bug; test=develop

* fix ptb author bug; test=develop

* Update mnist_dygraph.py

fix bug

* add muti card support for se_resnext

* add some description to readme.md

* add ce for cyclegan

* fix code style

* add ce for ptb_lm