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
@daquexian I‘m using grid sample op in pytorch, when converted to onnx, I write a custom op like: this. is there any way to add custom op in onnx_simpiler?
@ahangchen Thanks for your question. onnx-simplifier v0.3.0 allows you add custom onnx op through onnxruntime plugin. If your custom op is implemented as an onnxruntime plugin, you can upgrade your onnxsim version and pass --custom-lib the_path_of_the_lib when running onnxsim
Hi,
I am seeing:
onnx.onnx_cpp2py_export.checker.ValidationError: No Op registered for max_unpool2d with domain_version of 9
==> Context: Bad node spec: input: "276" input: "281" input: "284" output: "285" op_type: "max_unpool2d"
The model is plain pytorch segnet
The text was updated successfully, but these errors were encountered: