-
Notifications
You must be signed in to change notification settings - Fork 3
Open
Labels
enhancementNew feature or requestNew feature or request
Description
For example, when treevalue
is used in DI-treetensor
, some tensors may have the same shape (all dimensions are the same) and partly-same shape (e.g. 1st dimension is the same).
Maybe a constraint should be provided, to simplify the result, such as
from treevalue import TreeValue
import torch
t = TreeValue({'a': torch.randn(2, 3), 'b': {'x': torch.randn(2, 4)}})
t.shape[0] # should be a tree in current treevalue
# define some constraint
t.shape[0] # should be 2 after defining
Metadata
Metadata
Assignees
Labels
enhancementNew feature or requestNew feature or request