* fix fc padding during fusion, test=develop

* fix optim model inference after SaveOptimModel, test=develop

* fix communicator when breaking under PyReader mode, test=develop
* revert some vlog level to 0, test=develop

add lookup_table_dequant_op

As model fails when enable int8 quant, so disable allocate memory in cpu
for small variable.

* change the ci trt from version 5. to 6.0

* paddle-trt dynamic shape support init

* conv+bias or conv+bn dynamic shape support
test=develop

* modity trt engine opconvert
test=develop

* fix ci error
test=develop

* refine the profiler print test=develop

* polish detail implement of data loader, test=develop

* solve coverage ci problem, test=develop

fix concat_mkldnn op when encounter extreame conditions.

* reduce default attrs for dynamic graph, test=develop

* add some explanations for explicit attr, test=develop

* tweak explicit attr comments, test=develop

test=develop

* fix loaded program load bug; test=develop

* first version

* speed backward engin; test=develop

* remove useless code; test=develop

* reconvery io.py; test=develop

* remove useless code; test=develop

* remove useless code; test=develop

* add recorded cuda memory apis, fix typo, test=develop

* add more ut, test=develop

* follow comments, test=develop

* fix py35 incompatible issues, test=develop

* encapsulate the PaddleTensorToLoDTensor, test=develop

* serialize the pd_tensor, test=develop

* serialize tensors to file, test=develop

* add fluid.device_guard to specify the device type for Op

* change the function in op_teller, test=develop

* correct the commit-id, test=develop

* update ScopeBufferedSSAGraphExecutor&AsyncSSAGraphExecutor&ThreadedSSAGraphExecutor&FastThreadedSSAGraphExecutor&ParallelSSAGraphExecutor&ParallelExecutor for fetching unmerged results.

* add the unit test for fetch_unmerged.

* update ut for multi-card and multi-cpu.

* add the error message and the user suggestion in FetchOpHandle. test=develop

* add lodtensor share memory & serialization, test=develop

* fix windows compile error, test=develop

* deal vartype pickle & fix unittest matching error message, test=develop

* update timeout variable name, test=develop

* refactor memory map implement, test=develop

* clear mmap file discripter when exit unexpectedly, test=develop

* remove the child process fd in advance, test=develop

* remove mmap fds after Queue.put in child process, test=develop

* add hard unittests for register exit func, test=develop

* fix python2 compatibility problem in unittest, test=develop

* fix exception unittest error, test=develop

* polish code based review comment, test=develop

* fix inference c api lod, test=develop

* fix capi lod problem and enrich tests, test=develop

* delete useless header files and alter const_cast, test=develop

* add profiler_help.h to refine the code test=develop

* user can call dataset.set_download_cmd to set its customized download cmd
* add UT to cover this scenario

* Add the codegen and auto fusion for sum Op  in fusion group

* Refine adam op, test=develop

* Fuse kernels together to reduce cpu time.

* Refine paddle enforce, test=develop

* Remove some comments, test=develop

* Refine code,test=develop

* Refine cuda kernel, test=develop

* Refine code according to comments, test=develop

* Correct CPU gradients of the argsort op, form a network to test its forward and backward process, test=develop

* fix dynamic threshold error in test_argsort_op, test=develop

As the title

* refine pointer and c-api prototype, test=develop

* fix new c api profile bug, test=develop

* add unit tests, test=develop

* support cond in clone, test=develop

* refine code, test=develop

* refine code, test=develop

* follow comments, test=develop

* refine code, test=develop

* add framework overhead ratio, test=develop

* print GpuMemcpy overhead, test=develop

* register fp16 for assign op, test=develop

* add op test for fp16, test=develop

* Add two types of Metric Calculator: MultiTaskCalculator & CmatchRankCalculator.
* Add a config for DynamicAdjustChannelNum function to denote whether we will discard the remaining instances when they are not be distributed evenly.
* Remove CPU code in Pull/PushSparse and we will add it back when testing it fully.
* Fix some known issues: such as copying persistable vars after one epoch running.