It would be great if we could setup the training/validation scripts such that they are easy to use on new datasets.
There are likely some hardcoded solutions that needs to be fixed, but I believe most of the stuff is set through a configuration file? If that's the case, it shouldn't take that long.
It would be great if we could setup the training/validation scripts such that they are easy to use on new datasets.
There are likely some hardcoded solutions that needs to be fixed, but I believe most of the stuff is set through a configuration file? If that's the case, it shouldn't take that long.