Skip to content

2825 add hints for norm_name in dynunet_blocks.py#2826

Closed
yiheng-wang-nv wants to merge 3 commits into
Project-MONAI:devfrom
yiheng-wang-nv:2825-add-default-for-dynunet-block
Closed

2825 add hints for norm_name in dynunet_blocks.py#2826
yiheng-wang-nv wants to merge 3 commits into
Project-MONAI:devfrom
yiheng-wang-nv:2825-add-default-for-dynunet-block

Conversation

@yiheng-wang-nv

Copy link
Copy Markdown
Contributor

Signed-off-by: Yiheng Wang vennw@nvidia.com

Fixes #2825 .

Description

#2166 brings a breaking change for blocks in dynunet_block.py, before that commit, when using instance or group norm, affine=True is automatically used, but now this default setting is removed.

This PR adds some hints in the corresponding blocks, which is helpful for users to keep the consistency with old versions (0.5.3).

Status

Ready

Types of changes

  • Non-breaking change (fix or new feature that would not break existing functionality).
  • Breaking change (fix or new feature that would cause existing functionality to change).
  • New tests added to cover the changes.
  • Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
  • Quick tests passed locally by running ./runtests.sh --quick --unittests.
  • In-line docstrings updated.
  • Documentation updated, tested make html command in the docs/ folder.

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
@ahatamiz ahatamiz enabled auto-merge (squash) August 24, 2021 05:16
@ahatamiz ahatamiz disabled auto-merge August 24, 2021 06:50
@ahatamiz ahatamiz enabled auto-merge (squash) August 24, 2021 07:04
auto-merge was automatically disabled August 26, 2021 02:42

Pull request was closed

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.

Model Weights Saved in 0.5.3 may be incompatible with 0.6.0

1 participant