Problems encountered while compiling the onnx model

[Zeus1116]

I created an nnsmith environment using Conda and installed the required libraries according to the installation tutorial "python3- m pip install" nnsmith [torch, onnx, tvm, onnxruntime] "-- upgrade" in cli.md. I can generate the onnx model normally, but I encountered an issue when trying to debug this model locally.
When I used the command "nnsmith.model_exec model.type=onnx backend.type=onnxruntime model.path=nnsmith_output/model.onnx", an AttributeError has occurred
module 'onnx' has no attribute '_serialize'

This seems to be an issue with the onnx version, so I checked the onnx version and found that the current onnx version is 1.15.0, then I reinstalled version 1.14.0 of onnx, then I run the same command again, the onnx model will compile and run smoothly.

[Zeus1116]

I entered the tvm environment to reinstall the required configuration for nnsmith and ran the command "nnsmith. model_exec model. type=onnx"\

Backend. type=onnxruntime\

Model. path=nnsmith_ Output/model.onnx\

Cmp. with='{type: tvm, optmax: true, target: CPU}' 'conducted differential testing, but encountered an error as shown in the above figure. Can someone help me identify the cause.

[ganler]

Thanks for reporting the issue. Looks like you are right, onnx updated their API in 1.15.0 to load_from_string. Please bear a bit by downgrading the onnx versions for now before I brought up a fix.

[ganler]

I entered the tvm environment to reinstall the required configuration for nnsmith and ran the command "nnsmith. model_exec model. type=onnx"\

Backend. type=onnxruntime\

Model. path=nnsmith_ Output/model.onnx\

Cmp. with='{type: tvm, optmax: true, target: CPU}' 'conducted differential testing, but encountered an error as shown in the above figure. Can someone help me identify the cause.

Well, this is a TVM problem and nnsmith cannot help here.

More specifically, the pre-built TVM binary is compiled using a GLibC version which is newer than what your OS has right now. You can either recompile TVM locally to use your local glibc or install a newer glibc in conda.

[Zeus1116]

I entered the tvm environment to reinstall the required configuration for nnsmith and ran the command "nnsmith. model_exec model. type=onnx"
Backend. type=onnxruntime
Model. path=nnsmith_ Output/model.onnx
Cmp. with='{type: tvm, optmax: true, target: CPU}' 'conducted differential testing, but encountered an error as shown in the above figure. Can someone help me identify the cause.

Well, this is a TVM problem and nnsmith cannot help here.

More specifically, the pre-built TVM binary is compiled using a GLibC version which is newer than what your OS has right now. You can either recompile TVM locally to use your local glibc or install a newer glibc in conda.

I understand. Thanks for your help.