search

NVIDIA-AI-IOT / tf_trt_models

TensorFlow models accelerated with NVIDIA TensorRT
BSD 3-Clause "New" or "Revised" License
681 stars 245 forks source link

Import error: cannot import 'pipeline_pb2' #48

Open RogerAylagas opened 5 years ago

RogerAylagas commented 5 years ago

Hi, I am running on Jetson TX2. I have followed the setup instructions and after finishing I have copied your jupyter notebook example code to a python script. While trying to exectue it the following error occurs:

Traceback (most recent call last): File "run.py", line 14, in from tf_trt_models.detection import download_detection_model, build_detection_graph File "/home/nvidia/.local/lib/python3.5/site-packages/tf_trt_models-0.0-py3.5.egg/tf_trt_models/detection.py", line 1, in ImportError: cannot import name 'pipeline_pb2'

Can anyone help me with this? Thank you

RogerAylagas commented 5 years ago

I noticed that the folder /home/nvidia/.local/lib/python3.5/site-packages/tf_trt_models-0.0-py3.5.egg is not there. Why?

kongkongthu commented 4 years ago

I met the same question like yours, have you solved it? If so, can you show how you solved it.

jkjung-avt commented 4 years ago

Did you run 'install.sh'? I think the problem is caused by not compiling the '.proto' files with 'protoc'. More specifically, that should be done by this line of command in 'install.sh':

https://github.com/NVIDIA-AI-IOT/tf_trt_models/blob/master/install.sh#L26

kongkongthu commented 4 years ago

thank you very much, I forgot the install step.