[GitHub] [incubator-mxnet] DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors

2019-08-16 Thread GitBox
DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors URL: https://github.com/apache/incubator-mxnet/pull/15922#issuecomment-522169267 Understandable. It's really only a lot of cut-and-paste, but I'll reach out to someone else. --

[GitHub] [incubator-mxnet] DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors

2019-08-16 Thread GitBox
DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors URL: https://github.com/apache/incubator-mxnet/pull/15922#issuecomment-52216 Hey, can you weigh in on this PR? With the bundled speed improvement of test_shuffle, it's a major step in stabilizing t

[GitHub] [incubator-mxnet] DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors

2019-08-16 Thread GitBox
DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors URL: https://github.com/apache/incubator-mxnet/pull/15922#issuecomment-522040068 @larroy @marcoabreu Do you understand why the CI doesn't show here as completely passing? The 'miscellaneous' part also

[GitHub] [incubator-mxnet] DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors

2019-08-15 Thread GitBox
DickJC123 commented on issue #15922: Refactor for windows CI 'out of heap space' errors URL: https://github.com/apache/incubator-mxnet/pull/15922#issuecomment-521865392 I reviewed the original laop_6 failure in one of my other PR CI runs and found it was not related to tolerances. Hence r