Open thebigbai opened 5 years ago
Problem solved.
I've tried a lot of combinations, eg. cuda(9.0,10.0,10.1)+cudnn(7.5,7.6)+TRT(5.1.5, 5.1.2,5.0.2). I found if I install tensorflow by "conda install ", this problem is always occured.
INFO: tensorflow: Running against TensorRT version 0.0.0
Then, I changed my tensorflow installation to "pip install", problem solved. I don't know why, but solved.
Cheers,
Bigbai
Dear Ardian,
Thanks for your work. However I met some issues when try to run your project.
When I ran the optimizing of YOLOv3, there was not much improvement. The tensorRT model is not right. There are some info as following:
it seems there is no TRTEngineOp in the model:
Here is my environment : Ubuntu 16.04, tensorflow-gpu 13.1, cuda 10.1, cudnn 7.5 and tensorRT 5.1.5.0.
All the imported package is right, such as: import tensorrt import tensorflow.contrib.tensorrt as trt
Could you help me with this issue? Is my environment wrong?
Much thanks,
Bigbai