Skip to content

Skip key normalization in _resolve_indexing_state#2371

Open
brownbaerchen wants to merge 1 commit into
914_adv-indexing-outshape-outsplitfrom
914_skip_key_normalization
Open

Skip key normalization in _resolve_indexing_state#2371
brownbaerchen wants to merge 1 commit into
914_adv-indexing-outshape-outsplitfrom
914_skip_key_normalization

Conversation

@brownbaerchen

Copy link
Copy Markdown
Collaborator

The tests pass (locally) without this, so maybe we can remove this.
Not casting DNDarray to Tensor will be helpful in casting Tensor to DNDarray...

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Pull request overview

This PR removes the “key normalization” step in heat.core.dndarray._resolve_indexing_state(), aiming to avoid coercing local DNDarray index keys into torch.Tensor (and torch.int64) during indexing resolution.

Changes:

  • Removed pre-processing that coerced local (non-distributed) DNDarray index keys (and tuple/list components) to torch.int64 tensors before subsequent indexing logic.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread heat/core/dndarray.py
for k in key
)

# 1D boolean mask resolution
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: Todo

Development

Successfully merging this pull request may close these issues.

2 participants