Skip to content

Missing decomp for ops.aten.linalg_vector_norm #150207

@guangy10

Description

@guangy10

🐛 Describe the bug

This ops is heavily used in many vision transformers models (convnextv2, swiftformer, swinv2), lowering those models to ExecuTorch will fail because this op is neither in core aten opset or having a decomp rule.

We should consider adding a decomp rule for it to unblock

Versions

trunk

cc @jianyuh @nikitaved @pearu @mruberry @walterddr @xwang233 @lezcano @SherlockNoMad

Metadata

Metadata

Assignees

Labels

module: decompositionsTopics related to decomposition (excluding PrimTorch)module: linear algebraIssues related to specialized linear algebra operations in PyTorch; includes matrix multiply matmultriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions