Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

update: Update MSFT Contrib Ops in OVEP #521

Open
wants to merge 1 commit into
base: ovep-develop-lnl-1.4
Choose a base branch
from

Conversation

ankitm3k
Copy link

Description

This PR Updates the latest OV supported Contrib Ops in OVEP

@MayureshV1
Copy link

@ankitm3k, Can you please test a few models with these custom ops and ensure only the backends/ devices that are functional are called out as supported.

@@ -95,8 +95,10 @@ std::vector<SupportedOp> supported_op_mode = {
{"Atan", V_2020_4, {"CPU", "GPU"}},
{"Atanh", V_2020_4, {"CPU"}},
{"Atanh", V_2022_1, {"GPU"}},
{"Attention", V_2024_5, {"CPU","GPU","NPU"}},

Choose a reason for hiding this comment

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

@ankitm3k
Currently, we do not have specific operator mapping for the NPU. Instead, we rely on CPU operator support. If model compilation fails on the NPU, it automatically falls back to the CPU. As a result, graph partitioning is determined based on the operator support available on the CPU. So NPU can be removed from all the ops.

You can see this mapping in the constructor of get capability.
[https://github.com/intel/onnxruntime/blob/0981bbf4ca4af4d7216299f15de784f19ce6123a/onnxruntime/core/providers/openvino/ov_versions/capability.cc#L34]

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.

3 participants