Is there a way to select rows based on a loose distinct?

Question

I have a dataset with a lot of replicated rows, and I want to make a dataset with no replications. Date and time are the main ways of distinguishing between distinct and similar