1

Running latest WinMLRunning but getting error:

No suitable kernel definition found for op Add (node InceptionV1/InceptionV1/Conv2d_1a_7x7/BatchNorm/FusedBatchNorm)

I converted custom inference frozen Inception model to ONNX using VS or mmdnn. I'm getting similar error when writing new Win32 WinML desktop application from scratch. 

OS: 17758 insider preview
SDK : 17754 insider preview
VS: 15.8.4

WinML Runner
GPU: Intel(R) UHD Graphics 620

=================================================================
Name: tf2onnx
Author: tf2onnx
Version: 9223372036854775807
Domain:
Description:
Path: test3.onnx
Support FP16: false

Input Feature Info:
Name: input:0
Feature Kind: Float

Output Feature Info:
Name: InceptionV1/Logits/Predictions/Reshape_1:0
Feature Kind: Float

=================================================================

Loading model...[SUCCESS]
Creating session [FAILED]
No suitable kernel definition found for op Add (node InceptionV1/InceptionV1/Conv2d_1a_7x7/BatchNorm/FusedBatchNorm)

=================================================================
Name: tf2onnx
Author: tf2onnx
Version: 9223372036854775807
Domain:
Description:
Path: test3.onnx
Support FP16: false

Input Feature Info:
Name: input:0
Feature Kind: Float

Output Feature Info:
Name: InceptionV1/Logits/Predictions/Reshape_1:0
Feature Kind: Float

=================================================================

Loading model...[SUCCESS]
Creating session [FAILED]
No suitable kernel definition found for op Add (node InceptionV1/InceptionV1/Conv2d_1a_7x7/BatchNorm/FusedBatchNorm)
michael_heath
  • 5,262
  • 2
  • 12
  • 22
ms2579
  • 11
  • 2

1 Answers1

3

Please make sure that your operator set version is up to date. To check the version, open the model with Netron, click the input node and check the Imports field.

We want: ai.onnx v7 for ONNX Operator Set Version 7 which is the minimum version Windows Machine Learning API supports.

Try using the tf2onnx converter tool to convert from tensorflow. You may also need to update your ONNX python package. We support ONNX 1.2.2 or higher: https://github.com/onnx/tensorflow-onnx

This issue was resolved here: https://github.com/Microsoft/Windows-Machine-Learning/issues/50

Brian Martin
  • 80
  • 8
Rylai
  • 116
  • 6