I’ve been wracking my brain trying to find the python api UFF package and I’ve come to the conclusion that it’s not included in the nvidia docker container:
However a .whl file is included in the tensorRT5.0 release via https://developer.nvidia.com/nvidia-tensorrt-5x-download .
Someone should probably update the docker image so that the uff python package is pre-installed on the next release as following some of the examples are impossible without it
(e.g. /workspace/tensorrt/samples/python/end_to_end_tensorflow_mnist/)
1 Like
NVES
February 6, 2019, 7:57pm
2
Hello, are you referring to the uff-0.6.3-py2.py3-none-any.whl?
Selecting previously unselected package graphsurgeon-tf.
(Reading database ... 18014 files and directories currently installed.)
Preparing to unpack .../graphsurgeon-tf_5.0.2-1+cuda10.0_amd64.deb ...
Unpacking graphsurgeon-tf (5.0.2-1+cuda10.0) ...
Selecting previously unselected package uff-converter-tf.
Preparing to unpack .../uff-converter-tf_5.0.2-1+cuda10.0_amd64.deb ...
Unpacking uff-converter-tf (5.0.2-1+cuda10.0) ...
Setting up graphsurgeon-tf (5.0.2-1+cuda10.0) ...
Setting up uff-converter-tf (5.0.2-1+cuda10.0) ...
Please reference the instructions when the container starts:
=====================
== NVIDIA TensorRT ==
=====================
NVIDIA Release 18.12 (build 877198)
NVIDIA TensorRT 5.0.2 (c) 2016-2018, NVIDIA CORPORATION. All rights reserved.
Container image (c) 2018, NVIDIA CORPORATION. All rights reserved.
https://developer.nvidia.com/tensorrt
To install Python sample dependencies, run /opt/tensorrt/python/python_setup.sh
1 Like