microsoft / onnxruntime

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

Home Page:https://onnxruntime.ai

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Type Error: Type 'tensor(int64)' of input parameter (16586) of operator (Clip) in node (Clip_13292) is invalid.

lix19937 opened this issue · comments

Describe the issue

Use ort to infer, then raise error:

Type Error: Type 'tensor(int64)' of input parameter (16586) of operator (Clip) in node (Clip_13292) is invalid.

To reproduce

onnx                      1.15.0
onnx-graphsurgeon         0.3.21
onnx-simplifier           0.4.36
onnxruntime               1.17.1

Urgency

No response

Platform

Linux

OS Version

ubuntu20.04

ONNX Runtime Installation

Built from Source

ONNX Runtime Version or Commit ID

1.15.0

ONNX Runtime API

Python

Architecture

X64

Execution Provider

Default CPU

Execution Provider Library Version

No response

onnxruntime not support torch.clamp ?

Clip should be supported with int64 input.
https://github.com/microsoft/onnxruntime/blob/v1.17.1/docs/OperatorKernels.md#cpuexecutionprovider

What ONNX opset version is the model using?

What ONNX opset version is the model using?

opset_version in torch.onnx.export is 11, when I update to 13, no error.