Reconstruct. This should only be used When you have a single dtype for all columns. If not, you can end up with dtype object for all columns and changing them again involves much more dictionary work. df.columns = new wherever new is the listing of new columns names is as https://stack-ball-game-strategy98528.blogminds.com/stack-ball-free-online-game-things-to-know-before-you-buy-33455148