I trained the model using TAO toolkit 5.2.0 and it looks like I have to both run trtexec and pointpillar inference under the docker http://nvcr.io/nvidia/tao/tao-toolkit:5.2.0-pyt2.1.0 so that things would just work.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
OK, yes, suggest to do that inside the same docker.