You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
I've tried to copy the HelloPhi example for loading and calling phi-3.5 in DirectML for my application. When I get to the generator.ComputeLogits() part, I get a ton (e.g. hundreds) of the following memory errors:
Exception thrown at 0x00007FFFF9ED6D9A in <MyApplicationsName>.exe: Microsoft C++ exception: _com_error at memory location 0x0000006A74D6D680.
After a few minutes, the inference engine starts generating the text.
To Reproduce
Steps to reproduce the behavior:
Copy the DirectML version of the HelloPhi example as a new Blank App, Packaged with Windows Application Packing Project (WinUI 3 in Desktop) app with framework net8.0-windows10.0.19041.0 (n.b. I'm using WindowsSdkPackageVersion 10.0.19041.38).
Download and convert phi-3.5 to DirectML (or rely on the onnx dml verison at microsoft/Phi-3-mini-128k-instruct-onnx on huggingface - I've tried both models).
Run the app with the downloaded model (making sure to break when c++ exceptions are thrown.
See error
Expected behavior
No memory errors to show and to start generating text immediately (rather than after 3-5 minutes of the memory exceptions).
Screenshots
Desktop (please complete the following information):
OS: Windows 11
The text was updated successfully, but these errors were encountered:
Quick follow-up. I tried updating to the most recent Windows SDK (10.0.26100.0) and ran the app again. Now I'm getting a slightly different error (coming out of D3D12Core rather than KernelBase).
I’ve been in touch with Microsoft developers and we’ve pinpointed that this
is due to our packaging the build separately in our project. However, root
cause is still unclear.
Describe the bug
I've tried to copy the HelloPhi example for loading and calling phi-3.5 in DirectML for my application. When I get to the generator.ComputeLogits() part, I get a ton (e.g. hundreds) of the following memory errors:
After a few minutes, the inference engine starts generating the text.
To Reproduce
Steps to reproduce the behavior:
Blank App, Packaged with Windows Application Packing Project (WinUI 3 in Desktop)
app with framework net8.0-windows10.0.19041.0 (n.b. I'm using WindowsSdkPackageVersion 10.0.19041.38).Expected behavior
No memory errors to show and to start generating text immediately (rather than after 3-5 minutes of the memory exceptions).
Screenshots
Desktop (please complete the following information):
The text was updated successfully, but these errors were encountered: