* Add pad2d op.

* Add unitest and python api.

* Fix cuda op kernel.

* Fix python api.

* Fix python api.

* Update API.spec.

* Fix python api

* fix dist train reduce mode

* fix previous fix

* Add unit test for rpn_target_assign API.

* Add generate_proposal_labels for Faster-RCNN.

* Enhance the function of fused_elementwise_activation_op

* enhance unit test

* Clean Code And Add Doc

* Add compound functors

* Fix doc and enhance unit test

* define Dx and Dy for d_binary_func

* add mul_scale

* add mul_scale

* add elementwise_mul

* code refine

* code refine

* add doc

* add  AsIntermediate

* add test for dist and memopt

* update transformer too

* Add proposals generation operator for Faster-RCNN.

* Polish PrintOp

* Polish PrintOp

* Polish PrintOp

* Refine test_print_op

* Improve fake_dequantize_op.
* Follow comments.

* add ScaleOpVarTypeInference for scale op

* scale op support scale selected rows

* optimize code

* use FindVar

* use FindVarRecursive in ScaleOpVarTypeInference

* refine dist train RPC deps

* clean up

* clean up

* fix ut

* remove input for fetch_barrier

* follow comments

fix a few other things

* Support int8 type.

* Add pad_constant_batch_size_like

* refine pad_op

* optimize memory

This reverts commit b2df1700.

This reverts commit f5d5d7b2.

* Add Python Callstacks when Op::Run error

* Skip op with sub-block

* refactor: refine callstack info's format

* Reshape only support matrix

* Polish Python code

* Fix UT

* Fix Py3