Please do! Maybe the paper is technically enough to reproduce the results but if other researchers can start from a working example, they can both verify your r

Please do! Maybe the paper is technically enough to reproduce the results but if other researchers can start from a working example, they can both verify your results and extend them with more original research far faster. While any published code receives some nitpicking and bikeshedding, most academic code is terrible so unless you literally use random joke/meme variable names as your only 'documentation' (I wish I were joking) you're not going to look bad to anyone who matters.