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

add lovasz loss, mixed loss, road extration dataset #825

Merged
merged 10 commits into from
Feb 22, 2021

Conversation

LutaoChu
Copy link
Contributor

@LutaoChu LutaoChu commented Feb 3, 2021

1 新增lovasz softmax loss和lovasz hinge loss
2 新增MixedLoss类,在不改动组网代码下实现混合loss训练
3 新增mini deep globe road extraction二分类数据集

Copy link
Collaborator

@michaelowenliu michaelowenliu left a comment

Choose a reason for hiding this comment

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

已提review意见

backbone:
type: HRNet_W18
pretrained: https://bj.bcebos.com/paddleseg/dygraph/hrnet_w18_ssld.tar.gz
num_classes: 19
Copy link
Collaborator

Choose a reason for hiding this comment

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

model的num_classes去掉

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done

backbone:
type: HRNet_W18
pretrained: https://bj.bcebos.com/paddleseg/dygraph/hrnet_w18_ssld.tar.gz
num_classes: 19
Copy link
Collaborator

Choose a reason for hiding this comment

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

去掉num_classes

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done

backbone:
type: HRNet_W18
pretrained: https://bj.bcebos.com/paddleseg/dygraph/hrnet_w18_ssld.tar.gz
num_classes: 19
Copy link
Collaborator

Choose a reason for hiding this comment

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

去掉num_classes

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done

@@ -211,7 +211,9 @@ def loss(self) -> dict:
if key == 'types':
self._losses['types'] = []
for item in args['types']:
item['ignore_index'] = self.train_dataset.ignore_index
if item['type'] != 'MixedLoss':
item[
Copy link
Collaborator

Choose a reason for hiding this comment

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

改下格式

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done

There are 800 images in the training set and 200 images in the validation set.

Args:
dataset_root (str, optional): The dataset directory. Default: None
Copy link
Collaborator

Choose a reason for hiding this comment

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

注释最后加句号

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done

Forward computation.

Args:
logits: [N, C, H, W] Tensor, logits at each prediction (between -\infty and +\infty)
Copy link
Collaborator

Choose a reason for hiding this comment

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

Seg注释规范,类型

Copy link
Contributor Author

Choose a reason for hiding this comment

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

done

@michaelowenliu michaelowenliu merged commit 038cde8 into PaddlePaddle:develop Feb 22, 2021
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