TensorRT: TypeError: pybind11::init(): factory function returned nullptr.

Hi,i am use tensorrt7.0.0.11 and cuda10.1 ubuntu16.04 pytorch1.3.1 python3.6.But i faced above problem when i was using it.It indices the problem from this line:

TRT_LOGGER = trt.Logger(trt.Logger.WARNING)
with trt.Builder(TRT_LOGGER) as builder, builder.create_network() as 
 network, trt.OnnxParser(network,TRT_LOGGER) as parser:  #<--- this line got above problem.

The code can be use in the tensorrt6.0.1.5,but got problem in the tensorrt7.0.0.11.

What’s problem of my code?

About this issue

Original URL
State: closed
Created 4 years ago
Comments: 25

Most upvoted comments

＠rmccorm4,Hi,i have tried your suggestion, and i change my cuda10.1 to cuda10.0, but it still has such problem,the whole problem is: [TensorRT] ERROR: CUDA initialization failure with error 35. Please check your CUDA installation: http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html Traceback (most recent call last): master/scripts/Pytorch2ONNX.py", line 53, in ONNX2Engine with trt.Builder(TRT_LOGGER) as builder: TypeError: pybind11::init(): factory function returned nullptr Can you help me? tkx.

chegnyanjun on Jan 6, 2020

Facing same issue with the the nvcr.io/nvidia/tensorrt:20.01-py2 container. If I run the container interactively, there is no issue.

copaah on Jul 2, 2020

I got this error while using the official tensorrt docker file from the nvidia container repository.

nvcr.io/nvidia/tensorrt:20.03-py3

WillJStone on May 12, 2020

So you guys did not fix it for over 2 years?

kopyl on Jan 6, 2024

TypeError: pybind11::init(): factory function returned nullptr I encountered the same error, has anyone already solved this problem

binbin-w on Jan 12, 2024

TypeError: pybind11::init(): factory function returned nullptr I encountered the same error, has anyone already solved this problem

So where is the solution to it?

bigmover on Jan 18, 2024

For all those facing this error with nvidia docker like nvcr.io/nvidia/tensorrt:20.01-py2 nvcr.io/nvidia/tensorrt:20.03-py3 try running the docker as nvidia-docker run instead of docker run. This solved the issue for me.

sksenthilkumar on Oct 1, 2020

To the guys who are facing this issue in docker container, make sure -

You have CUDA 10.2 with tensorRT 7+
You have attached gpu to your container. --gpus all

vaibhawvipul on Jun 8, 2020

Hi @chegnyanjun,

I don’t think there’s anything wrong with that code snippet, I use it all the time. However, I notice you’re using CUDA 10.1, and I believe TRT7 only supports 9.0, 10.0, and 10.2 per the support matrix: https://docs.nvidia.com/deeplearning/sdk/tensorrt-support-matrix/index.html

Please try again after changing to a supported CUDA version or use a container if you don’t want to change your host install

rmccorm4 on Jan 2, 2020