Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

[Vision] add test cases for flip, normalize, to_tensor #8919

Merged
merged 5 commits into from
Dec 12, 2017

Conversation

yzhliu
Copy link
Member

@yzhliu yzhliu commented Dec 2, 2017

@@ -184,26 +184,26 @@ def forward(self, x):
return image.imresize(x, *self._args)


class RandomHorizontalFlip(HybridBlock):
"""Randomly flip the input image horizontally with a probability
class RandomLeftRightFlip(HybridBlock):
Copy link
Contributor

Choose a reason for hiding this comment

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

FlipLeftRight?

f_expect,
np.random.uniform(0, 255, (300, 300, 3)).astype(dtype=np.uint8))

def test_random_top_bottom_flip():
Copy link
Contributor

Choose a reason for hiding this comment

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

How about run tests on non random versions (flip_left_right, adjust_xxx)

@yzhliu
Copy link
Member Author

yzhliu commented Dec 4, 2017

I'll add tests for brightness/contrast... once this pr is approved

@@ -184,26 +184,44 @@ def forward(self, x):
return image.imresize(x, *self._args)


class RandomHorizontalFlip(HybridBlock):
"""Randomly flip the input image horizontally with a probability
class FlipLeftRight(HybridBlock):
Copy link
Contributor

@piiswrong piiswrong Dec 9, 2017

Choose a reason for hiding this comment

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

we don't need this for now. It's unclear when you would want to always flip in a transform. We only need the functional implementation flip_left_right

Copy link
Member Author

Choose a reason for hiding this comment

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

fixed

@marcoabreu
Copy link
Contributor

Hello @Javelinjs, please rebase your PR.

@yzhliu
Copy link
Member Author

yzhliu commented Dec 12, 2017

@marcoabreu You mean rebase from master? Is there anything related to these operators?

@marcoabreu
Copy link
Contributor

Ah my bad, I didn't see you're merging into a different branch. Nevermind :)

@piiswrong piiswrong merged commit 7c92b98 into apache:vision Dec 12, 2017
piiswrong pushed a commit to piiswrong/mxnet that referenced this pull request Jan 19, 2018
* [vision] ut for to_tensor, normalize, flip

* [vision] fix flip

* [vision] flip name

* [vision] test non-random flip op

* remove transform.FlipXXXX
piiswrong pushed a commit that referenced this pull request Jan 22, 2018
* [vision] ut for to_tensor, normalize, flip

* [vision] fix flip

* [vision] flip name

* [vision] test non-random flip op

* remove transform.FlipXXXX
yuxiangw pushed a commit to yuxiangw/incubator-mxnet that referenced this pull request Jan 25, 2018
* [vision] ut for to_tensor, normalize, flip

* [vision] fix flip

* [vision] flip name

* [vision] test non-random flip op

* remove transform.FlipXXXX
rahul003 pushed a commit to rahul003/mxnet that referenced this pull request Jun 4, 2018
* [vision] ut for to_tensor, normalize, flip

* [vision] fix flip

* [vision] flip name

* [vision] test non-random flip op

* remove transform.FlipXXXX
zheng-da pushed a commit to zheng-da/incubator-mxnet that referenced this pull request Jun 28, 2018
* [vision] ut for to_tensor, normalize, flip

* [vision] fix flip

* [vision] flip name

* [vision] test non-random flip op

* remove transform.FlipXXXX
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants