Hi,
Working on jetson orin nx. Installed the JP6.1, the trt10 broke all my c++ TRT code, with no backward compatibility.
Is there a way to run a docker with TRT8 on the JP6.1 , where the container is accessible to the GPU and (hopefully) to the DLA ?
Or
Install TRT8 on the jp6.1 in parallel/instead of the trt10 ?
Thanks for the help !
Hi,
It’s more recommended to use a newer version instead of downgrading.
What kind of error do you meet?
You can find the info in our release note to update the usage if it is related to API changes.
Thanks.
Hi,
The trt10 broke all my c++ code, and there is very little, to say the least, tutorials/samples on how to migrate. A lot of functions used in the code are now deprecated AND not supported in trt10 and it is very confusing how to fix the code. So, I have no real choice but to downgrade …
Any idea how to bypass this ?
Thanks for the help !
Hi,
It’s recommended to update your source to be compatible with TensorRT 10 instead.
You can find some information from our GitHub:
For example, you can find below the changes we made for the sampleOnnxMNIST.
samples/sampleOnnxMNIST/sampleOnnxMNIST.cpp
Thanks.