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

Change the transposed conv2d initializer. #984

Merged
merged 2 commits into from
Jun 15, 2018

Conversation

qingqing01
Copy link
Collaborator

@qingqing01 qingqing01 commented Jun 13, 2018

Fix #985

@@ -48,14 +49,16 @@ class PyramidBox(object):
def __init__(self,
data_shape,
num_classes,
use_transposed_conv2d=True,
Copy link
Contributor

Choose a reason for hiding this comment

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

用户需要知道use_transposed_conv2d=False的时候用的是什么upsampling方法么?
upsampling_method='transposed_conv' or 'resize_bilinear'会不会更直观?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

@wanghaoshuang Thanks, I'll add comments later.

param_attr=w_attr,
bias_attr=False)
else:
conv_up = fluid.layers.resize_bilinear(
Copy link
Contributor

Choose a reason for hiding this comment

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

conv_up 改成 upsampling 会不会更好些?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Done.

Copy link
Contributor

@wanghaoshuang wanghaoshuang left a comment

Choose a reason for hiding this comment

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

这些小问题后续可以另提PR修改。

@qingqing01 qingqing01 merged commit f629ce0 into PaddlePaddle:develop Jun 15, 2018
@qingqing01 qingqing01 deleted the conv2d_transpose_init branch January 28, 2019 02:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants