Closed shawrby closed 3 years ago
Hello. I apologise for the late answer. For both files, the RAD tensor has been aggregated using a sum over the third axis. We applied a log transformation on the representation to generate the _processed files. The _numpy are another preprocessing using a max instead of the sum for the range-angle, and it is a dupplicate of the *_processed for range-Doppler. I will homogenise the name of the files. You can recover these processed views using the RAD tensors which have been released lately. I hope it will help. Best
Thank you for making the radar data set public.
I'm trying to develop an algorithm using CARRADA dataset. What is the difference between '_raw, _processed, and *_numpy' files?
Thanks in advance for your reply.