Skip to content

Conversation

@zsdonghao
Copy link
Member

No description provided.

A :class:`DeConv2dLayer` object.
"""

Copy link
Member

Choose a reason for hiding this comment

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

Please re-add the deprecation header, just above the init function

Copy link
Member Author

Choose a reason for hiding this comment

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

done

# return net_new
else:
raise Exception("please update TF > 1.3 or downgrade TL < 1.8.4")
# if batch_size is None:
Copy link
Member

Choose a reason for hiding this comment

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

You can remove this code if it is not supported anymore.

self.all_params.extend(new_variables)
# return net_new
else:
raise Exception("please update TF > 1.3 or downgrade TL < 1.8.4")
Copy link
Member

Choose a reason for hiding this comment

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

Maybe the use of RuntimeError or NotImplementedError would be more appropriate

n = tl.layers.DeConv2d(nin, n_filter=32, filter_size=(3, 3), strides=(2, 2), name='DeConv2d')
print(n)
shape = n.outputs.get_shape().as_list()
print(shape[1:])
Copy link
Member

@DEKHTIARJonathan DEKHTIARJonathan Apr 17, 2018

Choose a reason for hiding this comment

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

Tests now collide with PR: #494
Please adopt the unittests structure and remove prints from tests before commit.

@DEKHTIARJonathan DEKHTIARJonathan added this to the 1.8.5 milestone Apr 17, 2018
@tensorlayer tensorlayer deleted a comment Apr 17, 2018
@tensorlayer tensorlayer deleted a comment Apr 17, 2018
"""

logging.info("DeConv2d %s: n_filters:%s strides:%s pad:%s act:%s" % (name, str(n_filter), str(strides), padding, act.__name__))
@deprecated_alias(layer='prev_layer', end_support_version=1.9) # TODO remove this line for the 1.9 release
Copy link
Member

Choose a reason for hiding this comment

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

This class has a double deprecation, please change for
@deprecated_alias(layer='prev_layer', n_out_channel='n_filter', end_support_version=1.9)

@DEKHTIARJonathan
Copy link
Member

@zsdonghao thanks for your corrections, only one small issue left. Once it's corrected I'll merge the PR ;)

print(shape[1:])
# if (shape[1] != 200) or (shape[2] != 200) or (shape[3] != 32): # TODO: why [None None None 32] ?
if (shape[3] != 32):
if (shape[1:] != [200, 200, 32]):
Copy link
Member

Choose a reason for hiding this comment

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

You could have kept the tests that you implemented before. Do you need help on this ?

@tensorlayer tensorlayer deleted a comment Apr 17, 2018
@zsdonghao zsdonghao merged commit c226cb3 into master Apr 17, 2018
@zsdonghao zsdonghao deleted the update-deconv2d branch April 17, 2018 12:23
luomai pushed a commit that referenced this pull request Nov 21, 2018
* update deconv2d function-->class, support dilation

* fix review comment

* fix review comment

* fix yapf

* fix review commend
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.

3 participants