Trying TensorRT Docker to optimize YoloV3 on GTX 1050Ti

I am trying to optimize YoloV3 using TensorRT.

I came this post called
Have you Optimized your Deep Learning Model Before Deployment?

Used Enabling GPUs in the Container Runtime Ecosystem to install nvidia-docker2

Pulled the latest version of the docker image docker pull aminehy/tensorrt-opencv-python3:version-1.3 from

Ran this

$sudo docker run -it --rm -v $(pwd):/workspace --runtime=nvidia -w /workspace -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=unix$DISPLAY aminehy/tensorrt-opencv-python3:version-1.3

Got this error

== NVIDIA TensorRT ==

NVIDIA Release 19.05 (build 6392482)

NVIDIA TensorRT 5.1.5 (c) 2016-2019, NVIDIA CORPORATION.  All rights reserved.
Container image (c) 2019, NVIDIA CORPORATION.  All rights reserved.

To install Python sample dependencies, run /opt/tensorrt/python/

root@a38b20eeb740:/workspace# cd /opt/tensorrt/python/
root@a38b20eeb740:/opt/tensorrt/python# chmod +x 
root@a38b20eeb740:/opt/tensorrt/python# ./
Requirement already satisfied: Pillow in /usr/local/lib/python3.5/dist-packages (from -r /opt/tensorrt/samples/sampleSSD/requirements.txt (line 1)) (6.0.0)
WARNING: You are using pip version 19.2.1, however version 19.3.1 is available.
You should consider upgrading via the 'pip install --upgrade pip' command.
Ignoring torch: markers 'python_version == "3.7"' don't match your environment
Setting up graphsurgeon-tf (5.1.5-1+cuda10.1) ...
Setting up uff-converter-tf (5.1.5-1+cuda10.1) ...
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/usr/lib/python2.7/dist-packages/uff/", line 1, in <module>
    from uff import converters, model  # noqa
  File "/usr/lib/python2.7/dist-packages/uff/model/", line 1, in <module>
    from . import uff_pb2 as uff_pb  # noqa
  File "/usr/lib/python2.7/dist-packages/uff/model/", line 6, in <module>
    from google.protobuf.internal import enum_type_wrapper
ImportError: No module named google.protobuf.internal
chmod: cannot access '/bin/': No such file or directory

The post says it is for Jetson device don’t we have a similar method for GTX platform?

Did post a question on here to