SthPhoenix / InsightFace-REST

InsightFace REST API for easy deployment of face recognition services with TensorRT in Docker.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

cannot start container from ./deploy_trt.sh

Asher-1 opened this issue · comments

commented

ERROR: CUDA initialization failure with error 222. Please check your CUDA installation: http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html
mxnet version: 1.7.0
onnx version: 1.7.0
Traceback (most recent call last):
File "prepare_models.py", line 45, in
prepare_models()
File "prepare_models.py", line 40, in prepare_models
prepare_backend(model_name=model, backend_name=backend_name, im_size=max_size, force_fp16=force_fp16,
File "/app/modules/model_zoo/getter.py", line 127, in prepare_backend
convert_onnx(temp_onnx_model,
File "/app/modules/converters/onnx_to_trt.py", line 79, in convert_onnx
engine = _build_engine_onnx(input_onnx=onnx_obj,
File "/app/modules/converters/onnx_to_trt.py", line 25, in _build_engine_onnx
with trt.Builder(TRT_LOGGER) as builder,
TypeError: pybind11::init(): factory function returned nullptr
Starting InsightFace-REST using 1 workers.
[09:43:07] INFO - Reshaping ONNX inputs to: (1, 3, 640, 640)
mxnet version: 1.7.0
onnx version: 1.7.0

Hi, @Asher-1 ! Looks like you have older GPU drivers version, ensure that you have 460.x.x version drivers.

commented

thank you, that works for me on 460.x.x

Closed as resolved.