Skip to content

Time-distributed flexible UNet for 2D and 3D space#2

Open
MarkWrobelUCL wants to merge 1 commit intomainfrom
mark_dev
Open

Time-distributed flexible UNet for 2D and 3D space#2
MarkWrobelUCL wants to merge 1 commit intomainfrom
mark_dev

Conversation

@MarkWrobelUCL
Copy link
Copy Markdown
Contributor

No description provided.

@MarkWrobelUCL MarkWrobelUCL requested a review from aniraman1 March 25, 2026 16:46
Copy link
Copy Markdown
Collaborator

@aniraman1 aniraman1 left a comment

Choose a reason for hiding this comment

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

Looks good, all the code is very neat! I left some suggestions for moving the code to different parts of the repo

return tuple(kk // 2 for kk in k)


class TimeDistributed(nn.Module):
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

I would move all the classes to a separate file (like time_dist.py or something like that)

from ..layers.layer_util import get_image_layer, get_activation, standardize_spatial_factors, _same_padding_time
from ..layers.layer_util import MaxPoolTime, UpsampleTime, ConvTime, BatchNormTime, ConvTransTime

class ImageConvBlockTime(nn.Module):
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

I think the conv blocks and encoder/decoder probably belong in layers rather than models?

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