This repository has been archived by the owner on Nov 17, 2023. It is now read-only.
Sequence last fix #16156
Merged
Sequence last fix #16156
Conversation
This file contains 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
ChaiBapchya
force-pushed
the
sequence_last_fix
branch
from
e8883e4
to
1411a42
Compare
ChaiBapchya
force-pushed
the
sequence_last_fix
branch
from
0831d07
to
dfcd599
Compare
|
@mxnet-label-bot add [pr-awaiting-review] |
|
@access2rohit @apeforest take a look. Thanks. |
|
access2rohit
approved these changes
Sep 12, 2019
|
@ChaiBapchya Can you also run the full suite of tests again and paste the results here once they are done |
|
For C5.18xl instance (similar to the one used to run Nightly build) topk,argsort,sort - ERROR multinomial - fails due to memory shortage Upon running multinomial individually it passes |
|
Subsequent run (commenting the ones already tested) Individually all the error tests work |
access2rohit
pushed a commit
to access2rohit/incubator-mxnet
that referenced
this pull request
Sep 25, 2019
* seq last fix * index tensor to have int64 * fix dtypes * revert unnecessary changes * if seq len not passed, pass int64 dtype * dtype comment * use int32 or int64 as index dtype based on build flag * Trigger notification * Trigger notification * lint fix
access2rohit
pushed a commit
to access2rohit/incubator-mxnet
that referenced
this pull request
Sep 25, 2019
* seq last fix * index tensor to have int64 * fix dtypes * revert unnecessary changes * if seq len not passed, pass int64 dtype * dtype comment * use int32 or int64 as index dtype based on build flag * Trigger notification * Trigger notification * lint fix
access2rohit
pushed a commit
to access2rohit/incubator-mxnet
that referenced
this pull request
Sep 25, 2019
* seq last fix * index tensor to have int64 * fix dtypes * revert unnecessary changes * if seq len not passed, pass int64 dtype * dtype comment * use int32 or int64 as index dtype based on build flag * Trigger notification * Trigger notification * lint fix
larroy
pushed a commit
to larroy/mxnet
that referenced
this pull request
Sep 28, 2019
* seq last fix * index tensor to have int64 * fix dtypes * revert unnecessary changes * if seq len not passed, pass int64 dtype * dtype comment * use int32 or int64 as index dtype based on build flag * Trigger notification * Trigger notification * lint fix
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Description
Pass dtype to sequence_last operator.
Why?
Unless explicitly specified, default MXNet implementation turns NDArray into dtype=Float32
However, indices need to support int64 (>2**32)
Checklist
Essentials
Please feel free to remove inapplicable items for your PR.
Changes
Comments
Hopefully, for v2.0 of MXNet, we introduce API breaking change of ensuring NDArray for indices are either int32 or int64