test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

test=develop

- Added rough profiling

- Profiled maxpool itself

- First draft of max seqpool optimization (is_test added)

- Added unit tests to seqpool

- Cosmetic fixes

- Fix to UT of Seq pool

Disabled grad checking for sequence max pool when is_test is set to True

-Cosmetic fix to comment

test=develop

- Fix to GPU build

test=develop

- yet another GPU fix for sequence max pool

- Fix to comment

test=develop

- Change to API of sequence_pool

test=develop

- Yet another API spec change

test=develop

* fix nccl2 lars dist support

* put lars in momentum op

* add tests lars

* fix ci

* fix cpu kernel

* soft warning

* remove lars in test_recognize_digits.py

* move to another op

* add file

* update api.spec test=develop

* update test=develop

* fix api.spec test=develop

* wip

* wip, finish grad merge ops

* wip, finish graph build

* wip test running

* work on 1 gpu

* workable version

* update

* fix tests

* fuse broadcast op

* fix compile failed

* refine

* add batch merge test mnist

* fix CI test=develop

* fix build

* use independent bn params for batch merge test=develop

* update api.spec

* follow comments and for test

* wip

* refine tests test=develop

* follow comments test=develop

* remove startup bn modify test=develop

* follow comments test=develop

* fix merge test=develop

test=develop

close UT about dist ctr, will fix it later
test=develop

* fix graph num hang test=develop

* re-enable tests test=develop

* re-enable graph num check test=develop

* fix multi device pass role check test=develop

test=develop

test=develop

Fix transformer unittest

* disable async dist test

* update test=develop

test=develop

test=develop

1. Fix CUDA kernel when height is large than 2048.
2. Support input with more than 2D.
3. Fix unit test when k is large than 1.
4. Enhence unit testing.

test=develop

avoid the default one that usually overwrites manually set ones

test=develop