* change CUDA implementaion of uniform/gaussian OP

* fix unittest

* add pplayer

* add sharedlayerdesc

* add parallel_cross_entropy

* add grad for crossentropy

* fix cross entropy

* fix c_split bug

* fix utest

* add c_embedding for tensorparallel

* fix precision of mp

* fix bug of seed

* fix dp

* print group

* add clip/check

* add amp & clip grad in dygraph

* add logging

* add model parallel support in dygraph