Adding lrn op for ngraph engine#17189
Merged
tensor-tang merged 3 commits intoPaddlePaddle:developfrom May 8, 2019
Merged
Conversation
Contributor
Do we have plan to digger deeply why? Same with MKLDNN LRN. |
Contributor
|
@tensor-tang Sure, we can back to look at this one - before long. Right now it is impossible because we have other urgent issues. On the whole, this pull-request is really important for us, due to that CAPIs tests blow up. |
tensor-tang
approved these changes
May 8, 2019
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Added lrn op for ngraph engine which is needed for googlenet. A CreateConst helper function was added as well.
In the unittest, the tolerance was relaxed to 0.002, same as mkldnn test case. Otherwise I see the error as the following (don't quite understand the random decimal digits).
AssertionError: Output (MidOut) has diff at CPUPlace
514: Expect [[[[2.0005884 2.0006845 2.0005584 2.0006466 2.000742 ]
514: [2.0004916 2.0007987 2.0007715 2.0007942 2.0006254]
514: [2.0006313 2.0007539 2.0008295 2.0007706 2.0007606]
514: [2.000683 2.0005753 2.0004666 2.0005774 2.0008302]
514: [2.0005436 2.0009174 2.0009565 2.0007875 2.00066 ]]
514:
514: [[2.0005884 2.0006845 2.0005584 2.0006466 2.000742 ]
514: [2.0004916 2.0007987 2.0007715 2.0007942 2.0006254]
514: [2.0006313 2.0007539 2.0008295 2.0007706 2.0007606]
514: [2.000683 2.0005753 2.0004666 2.0005774 2.0008302]
514: [2.0005436 2.0009174 2.0009565 2.0007875 2.00066 ]]
514:
514: [[2.0005884 2.0006845 2.0005584 2.0006466 2.000742 ]
514: [2.0004916 2.0007987 2.0007715 2.0007942 2.0006254]
514: [2.0006313 2.0007539 2.0008295 2.0007706 2.0007606]
514: [2.000683 2.0005753 2.0004666 2.0005774 2.0008302]
514: [2.0005436 2.0009174 2.0009565 2.0007875 2.00066 ]]]
514:
514:
514: [[[2.0008764 2.0006783 2.0005012 2.0005872 2.0006566]
514: [2.0007496 2.000762 2.000739 2.0008848 2.0009556]
514: [2.0006292 2.0005717 2.0003912 2.0007153 2.0006144]
514: [2.0006511 2.0008154 2.0007868 2.0006576 2.0008616]
514: [2.0009568 2.000582 2.0004728 2.000677 2.0008805]]
514:
514: [[2.0008764 2.0006783 2.0005012 2.0005872 2.0006566]
514: [2.0007496 2.000762 2.000739 2.0008848 2.0009556]
514: [2.0006292 2.0005717 2.0003912 2.0007153 2.0006144]
514: [2.0006511 2.0008154 2.0007868 2.0006576 2.0008616]
514: [2.0009568 2.000582 2.0004728 2.000677 2.0008805]]
514:
514: [[2.0008764 2.0006783 2.0005012 2.0005872 2.0006566]
514: [2.0007496 2.000762 2.000739 2.0008848 2.0009556]
514: [2.0006292 2.0005717 2.0003912 2.0007153 2.0006144]
514: [2.0006511 2.0008154 2.0007868 2.0006576 2.0008616]
514: [2.0009568 2.000582 2.0004728 2.000677 2.0008805]]]]
514: But Got[[[[2. 2. 2. 2. 2.]
514: [2. 2. 2. 2. 2.]