Backbone of FBA. #215
Merged
Backbone of FBA. #215
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov Report
@@ Coverage Diff @@
## master #215 +/- ##
==========================================
- Coverage 82.27% 81.71% -0.57%
==========================================
Files 145 148 +3
Lines 6732 7055 +323
Branches 1004 1047 +43
==========================================
+ Hits 5539 5765 +226
- Misses 1086 1171 +85
- Partials 107 119 +12
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
innerlee
reviewed
Mar 1, 2021
| batch_norm=False, | ||
| norm_cfg=None, | ||
| act_cfg=dict(type='ReLU')): | ||
| super(FBADecoder, self).__init__() |
innerlee
reviewed
Mar 1, 2021
| x = torch.cat((x, conv_out[-6][:, :3], img, two_chan_trimap), 1) | ||
|
|
||
| output = self.conv_up4(x) | ||
| alpha = torch.clamp(output[:, 0][:, None], 0, 1) |
There was a problem hiding this comment.
[:, 0][:, None] -> [:, 0:1]
innerlee
reviewed
Mar 1, 2021
| @@ -0,0 +1,119 @@ | |||
| import torch | |||
There was a problem hiding this comment.
Give credit if the code is modified from the original code
innerlee
reviewed
Mar 1, 2021
| norm(256, self.batch_norm), nn.LeakyReLU()) | ||
|
|
||
| # Keep the batch_norm here in case we may need to modify something | ||
| if (self.batch_norm): |
innerlee
reviewed
Mar 1, 2021
| ConvWS2d(256 + 256, 256, kernel_size=3, padding=1, bias=True), | ||
| norm(256, self.batch_norm), nn.LeakyReLU()) | ||
|
|
||
| # Keep the batch_norm here in case we may need to modify something |
There was a problem hiding this comment.
in case we may need to modify something
What does this line of comment mean
innerlee
reviewed
Mar 1, 2021
| nn.Conv2d(16, 7, kernel_size=1, padding=0, bias=True)) | ||
|
|
||
| def init_weights(self, pretrained=None): | ||
| pass |
innerlee
reviewed
Mar 1, 2021
| def forward(self, inputs): | ||
| """Forward function. | ||
| Args: | ||
| inputs (dict): Output dict of FbaEncoder. |
innerlee
reviewed
Mar 1, 2021
|
|
||
| conv_out = inputs['conv_out'] | ||
| img = inputs['merged'] | ||
| two_chan_trimap = inputs['two_chan_trimap'] |
There was a problem hiding this comment.
two_chan_trimap -> two_channel_trimap
innerlee
reviewed
Mar 1, 2021
| nn.Sequential( | ||
| nn.AdaptiveAvgPool2d(scale), | ||
| ConvWS2d(2048, 256, kernel_size=1, bias=True), | ||
| norm(256, self.batch_norm), nn.LeakyReLU())) |
There was a problem hiding this comment.
Why norm_cfg and act_cfg are needed if all are hard-coded as batch_norm and LeakyReLU
innerlee
reviewed
Mar 1, 2021
|
|
||
|
|
||
| def norm(dim, bn=False): | ||
| if (bn): |
innerlee
reviewed
Mar 1, 2021
| return out | ||
|
|
||
|
|
||
| class ResNet(nn.Module): |
innerlee
reviewed
Mar 1, 2021
| } | ||
|
|
||
|
|
||
| def l_resnet50(pretrained=None, **kwargs): |
There was a problem hiding this comment.
The function name is cryptic
added 13 commits
March 2, 2021 11:40
innerlee
approved these changes
Mar 4, 2021
Yshuo-Li
pushed a commit
to Yshuo-Li/mmediting
that referenced
this pull request
Jul 15, 2022
* Backbone of FBA. * Init. * Doc string for forward. * Doc string of Init. * Modified API. * FBAencoder. * Tiny. * Doc string. * Decoder. * Init. * General Res. * Added test. * Added tests. * Added two_channel_trimap key. * Tiny. * Restore shape. * Tiny. * Added utis. May change. * Added FBA mattor. * Improved. * Tiny. * Modified. * Tiny. * Tiny. * Tiny. * Delete plugins. * Modified. * Tiny. * Mattor. * Tiny. * Tiny. * Postponed to next branch. * Tiny. * Update base_mattor.py Co-authored-by: lizz <innerlee@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.