added large tensor support for add_n and tests for more ops#16476
added large tensor support for add_n and tests for more ops#16476anirudh2290 merged 1 commit intoapache:masterfrom
Conversation
ChaiBapchya
left a comment
There was a problem hiding this comment.
LGTM with minor nitpicks.
|
@sxjscience @apeforest @anirudh2290 @zheng-da @pengzhao-intel This PR is ready for review |
| assert z[-1] == 2 | ||
| z = nd.minimum(x, 5) | ||
| assert z[0] == 3 | ||
| assert z[-1] == 3 |
There was a problem hiding this comment.
are the backward pass of any of these functions tested.
There was a problem hiding this comment.
Only forward passes are required by DGL. So only forward passes are tested.
There was a problem hiding this comment.
is there some large tensor docs in mxnet website. That should explicitly call out that the backward pass has not been verified with large tensor.
There was a problem hiding this comment.
I have to write that. Currently we don't have any docs explaining how to use large tensor support and caveats along with limitations and perf regressions.
014a137 to
1917af8
Compare
|
@anirudh2290 good to merge? |
1917af8 to
da721d8
Compare
|
@mxnet-label-bot add [pr-awaiting-merge] |
da721d8 to
c6ac555
Compare
Description
Added large tensor support for add_n and tests for:
load()
save()
add_n()
modulo()
maximum()
minimum()
Checklist
Essentials
Testing
Will run full suite of tests and paste here.