Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

[Performance regression] Revamp IntSet #3272 causing GluonCV SSD performance issue

See original GitHub issue

https://github.com/dmlc/tvm/pull/3272 is causing the similar issue used to happen in https://github.com/dmlc/tvm/issues/3097

The operator fused_strided_slice_greater_cast_strided_slice_zeros_like_add_add_add_add_add_ad_11203150218747419416_ is much slower due to mod operation not simplified:

placeholder[((((ax0.ax1.fused*4) + ax2) + -466036) % 16)]

While before this PR it is:

placeholder[((((ax0.ax1.fused*4) + ax2) + -4) % 16)]

@tqchen @wweic

Issue Analytics

State:
Created 4 years ago
Comments:19 (19 by maintainers)

Top GitHub Comments

1reaction

tqchencommented, Oct 8, 2019

@kevinthesun please check again now that the the integer simplification infra lands

1reaction

tqchencommented, Jul 9, 2019

Yes, introducing floordiv/mod might improve the perf further, but will need a few more PRs to change the division mode to take benefit of that. I would encourage us to separate the issue. If you can still isolate things that can be improved, we can dig further here

Top Results From Across the Web

[Performance] x86 gluoncv ssd performance regression #3097

kevinthesun mentioned this issue on Jun 30, 2019. [Performance regression] Revamp IntSet #3272 causing GluonCV SSD performance issue #3465.

05. Deep dive into SSD training: 3 tips to boost performance

In this article, we will dive deep into the details and introduce tricks that important for reproducing state-of-the-art performance.

gluoncv.model_zoo.ssd.ssd — gluoncv 0.11.0 documentation

Its length must be equals to the number of SSD output layers. steps : list of int Step size of anchor boxes in...

Detection — gluoncv 0.11.0 documentation

The following tables list pre-trained models for object detection and their performances with more details. Hint. Model attributes are coded in their names....

04. Train SSD on Pascal VOC dataset

This tutorial goes through the basic building blocks of object detection provided by GluonCV. Specifically, we show how to build a state-of-the-art Single ......