Fix flaky test poisson generator & test_negative_binomial_generator

[stu1130]

Description

Related to #14540
Fixes #13584 Fixes #13506
Adjust the default success_rate back to 0.2.
Originally it was 0.25 after #13498 it changed to 0.15.

passed all the previous failed seed.
ran both of the test case 10k times

MXNET_TEST_COUNT=10000 nosetests -v test_operator_gpu.test_poisson_generator
[INFO] Setting module np/mx/python random seeds, use MXNET_MODULE_SEED=1635133297 to reproduce.
test_operator_gpu.test_poisson_generator ... ok

----------------------------------------------------------------------
Ran 1 test in 20881.518s

OK

MXNET_TEST_COUNT=10000 nosetests -v test_operator_gpu.test_negative_binomial_generator
[INFO] Setting module np/mx/python random seeds, use MXNET_MODULE_SEED=636644805 to reproduce.
test_operator_gpu.test_negative_binomial_generator ... ok

----------------------------------------------------------------------
Ran 1 test in 26051.124s

OK

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

• The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage:
• Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
• Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
• Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
• Code is well-documented:
• For user-facing API changes, API doc string has been updated.
• For new C++ functions in header files, their functionalities and arguments are documented.
• For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
• Check the API doc at http:https://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
• To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

Comments

[piyushghai]

@stu1130 Thanks for fixing the flaky tests.

@mxnet-label-bot Add [CI, Flaky, pr-awaiting-review]

[piyushghai]

LGTM.

[piyushghai]

@stu1130 Can you look into the CI failures though ?

[larroy]

LGTM

[szha]

@stu1130 could you rebase to the latest master?

[stu1130]

@szha I have rebased to the latest master

commit 9822b17f36be6c6409b01371ca5aaafe6b204f2f (HEAD -> fix_test_poisson_generator, origin/fix_test_poisson_generator)
Author: stu1130 <*****@gmail.com>
Date:   Wed Mar 27 14:08:34 2019 -0700

    fix the flaky test

commit 8a64f606c2d9407db5ca0fe8e4dbfc61e45aee2b
Author: stu1130 <*****@gmail.com>
Date:   Wed Mar 27 10:12:34 2019 -0700

    set success_rate to 0.2

commit e2f5b47346e148c2376da7e6628750747f2d6a94 (upstream/master)
Author: JackieWu <[email protected]>
Date:   Wed Apr 3 01:53:02 2019 +0800

    Support SyncBatchNorm5D (#14542)

    * support SyncBatchNorm5D

    * fix
...