Comment by marvel_boy

Not bad, is just useless. ML canot distinguish between (a equal b) and (b equal a). For some tasks it is just an stocastic parrot.