test=document_preview

* update the description of name for floor and hard_sigmoid op

* update the description of name for floor op

* update conv2d 
conv3d
conv2d_transpose
conv3d_transpose
batch_norm
cos_sim
cross_entropy
npair_loss
dropout
img_conv_group cn doc

* refine square and squeeze name parameter
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* refine square api doc
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* fix train_from_dataset and infer_from_dataset
* fix DataFeedDesc, set_use_slot and desc

* fix concat doc
test=document_preview

* fix
test=document_preview

* fix concat attr name
test=document_preview

* fix hash doc
test=document_preview

* fix
test=document_preview

* fix
test=document_preview

* fix
test=document_preview

* fix hash attr name
test=document_preview

* fix cast doc
test=document_preview

* fix
test=document_preview

* fix
test=document_preview

* fix
test=document_preview

* fix
test=document_preview

* fix doc of pool2d pool3d. test=develop,test=document_preview

* fix doc according to reviews. test=develop,test=document_preview

* fix doc according to reviews from XiaoguangHu01 and raindrop2sea. test=develop,test=document_preview

* fix Varaible doc

* fix LOD to LoD

* fix LOD to LoD

* test=develop, refine doc

* test=develop, refine doc

* test=develop, refine doc and example code of Variable_cn.rst

* fix Print, is_empty, read_file op' cn doc test=develop

fix the doc of API:fluid.layers.py_func

According to the latest documentation standards：

* fix the document of API:fluid.Layers.sum

* fix the document of API:fluid.Layers.sum

* fix the document of API:fluid.Layers.sum

* fix the chinese doc for paddle test=develop
paddle.fluid.layers.create_global_var
paddle.fluid.layers.create_parameter
paddle.fluid.optimizer.FtrlOptimizer
paddle.fluid.optimizer.LarsMomentumOptimizer
paddle.fluid.optimizer.MomentumOptimizer
paddle.fluid.optimizer.RMSPropOptimizer
paddle.fluid.optimizer.SGDOptimizer

* Refine StaticRNN api doc

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* Refine StaticRNN api doc
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* fix 5 api_cn documents
test=document_preview

* modify io_map_readers

* Refine topk op api doc.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* refine top k api doc
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* polish decayed adagrad opt zh api doc

* change the note position

* delete return and return type

* polish details

* use whole optimizer name

* polish adamax optimizer zh doc

* polish details

* use whole optimizer name

* polish adam optimizer zh api doc

* polish details

* polish adagrad optimizer api zh doc

* polish details

* refine fluid.layers.one_hot

* update one_hot doc

* fix code error

* refine docs of CUDA and CPU places, test=develop

* refine docs of CUDA and CPU places, test=develop

* refine cuda_pinned_places, test=develop

* refine, test=develop

* refine, test=develop

* polish seven minimize api zh doc

* polish details

* enrich grad_clip param comment

* (4) modify reorder_lod_tensor_by_rank doc

* refine scope_guard_cn, test=develop

* refine scope_guard_cn, test=develop

* refine, test=develop

* refine, test=develop

update uniform_random_batch_size_like_cn.rst

* refine api docs name param

* add name param refine for ifelse,round,sqrt,sin,scatter,test=develop

* refine name according the new standard test=develop

* refine name according the new standard test=develop