Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

QSeparableConv1D and 2D #50

Closed
wants to merge 9 commits into from
Closed

QSeparableConv1D and 2D #50

wants to merge 9 commits into from

Conversation

vloncar
Copy link
Contributor

@vloncar vloncar commented Oct 16, 2020

This PR adds QSeparableConv1D and QSeparableConv2D. The existing QSeparableConv2D (which expands to QDepthwiseConv2D and 1x1 QConv2D) that is based on MobileNet is retained and renamed QMobileNetSeparableConv2D

@google-cla google-cla bot added the cla: yes contribution license agreement signed label Oct 16, 2020
@zhuangh
Copy link
Contributor

zhuangh commented Oct 24, 2020

cc @lishanok @danielemoro

@zhuangh zhuangh added the ready to pull to trigger copybara-service label Oct 24, 2020
@zhuangh zhuangh removed the ready to pull to trigger copybara-service label Nov 15, 2020
@zhuangh zhuangh added the ready to pull to trigger copybara-service label Nov 15, 2020
@zhuangh
Copy link
Contributor

zhuangh commented Nov 20, 2020

@lishanok could you check this PR and internal CL? thanks.

cc @danielemoro

@zhuangh zhuangh removed the ready to pull to trigger copybara-service label Dec 7, 2020
@zhuangh
Copy link
Contributor

zhuangh commented Dec 24, 2020

happy holidays! courtesy ping @lishanok and @danielemoro

@zhuangh
Copy link
Contributor

zhuangh commented Jan 5, 2021

PTAL @lishanok and @danielemoro

Copy link
Collaborator

@danielemoro danielemoro left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I believe this pull request is important since it properly implements QSeparableConv2D as a class and quantized layer instead of a function that simply creates other layers. However, we need to make sure that all parts of the QSeparableConv2D can be quantized.

qkeras/qconvolutional.py Show resolved Hide resolved
tests/qconvolutional_test.py Show resolved Hide resolved
tests/qconvolutional_test.py Outdated Show resolved Hide resolved
tests/qconvolutional_test.py Show resolved Hide resolved
@zhuangh zhuangh added the ready to pull to trigger copybara-service label Feb 2, 2021
qkeras-robot pushed a commit that referenced this pull request Feb 2, 2021
--
1847f67 by Vladimir Loncar <vloncar@users.noreply.github.com>:

Separable CNN implementation

--
bb717bc by Vladimir Loncar <vloncar@users.noreply.github.com>:

Add support for SeparableConv in utils

--
cc23ec4 by Vladimir Loncar <vloncar@users.noreply.github.com>:

Test case for separable conv

--
bb2e6c7 by Vladimir Loncar <vloncar@users.noreply.github.com>:

Support SeparableConv in AutoQ

--
72e2bd0 by Vladimir Loncar <vloncar@users.noreply.github.com>:

Update readme with new layers

--
5ab8ec0 by Vladimir Loncar <vloncar@users.noreply.github.com>:

Add a note about quantizing depthwise activations

COPYBARA_INTEGRATE_REVIEW=#50 from vloncar:sep_conv 5ab8ec0
PiperOrigin-RevId: 355256946
Change-Id: I106d43da286e519899425a83544ffa2e4afedb35
@zhuangh
Copy link
Contributor

zhuangh commented Feb 3, 2021

The test is actually not failed since we have it tested successfully from internal system. It was just due to travis-ci.org service failure.

@zhuangh
Copy link
Contributor

zhuangh commented Feb 4, 2021

closed this PR since it was merged cb046f9
Thanks @vloncar!

@zhuangh zhuangh closed this Feb 4, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
cla: yes contribution license agreement signed ready to pull to trigger copybara-service
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants