Sets

Unique values

A set is a collection where every value appears at most once. Sets are written with curly braces, like dictionaries, but contain single values rather than key-value pairs. If you add a value that is already there, nothing changes.

Sets are unordered: they don't keep items in the order you added them, so you can't use indexes like s[0]. When you need the items in order, for example to print them, use sorted(), which returns a sorted list.

elements = {"Fe", "Cr", "Ni", "Fe", "Cr"}print(len(elements))print(sorted(elements))

Removing duplicates from a list

Converting a list to a set with set() throws away duplicates. This is the quickest way to find the distinct values in a dataset.

structures = ["FCC", "BCC", "FCC", "HCP", "BCC", "FCC"]distinct = set(structures)print(f"{len(structures)} samples, {len(distinct)} structures")print(sorted(distinct))

Adding and removing items

add(x) puts one value into a set. remove(x) takes it out but raises a KeyError if it isn't there; discard(x) does the same without an error. Testing membership with in is very fast for sets, even when they are large.

alloying = {"Cr", "Ni"}alloying.add("Mo")alloying.add("Cr")          # already present: no changealloying.discard("W")       # not present: no errorprint(sorted(alloying))print("Mo" in alloying)

Set operations

Sets support the operations from mathematical set theory. They are a clear way to compare two groups: which items they share, which belong to only one, and which appear in either.

steel_304 = {"Fe", "Cr", "Ni", "C"}steel_316 = {"Fe", "Cr", "Ni", "Mo", "C"}print("Shared:", sorted(steel_304 & steel_316))print("Only in 316:", sorted(steel_316 - steel_304))print("All elements:", sorted(steel_304 | steel_316))

The molybdenum in 316 stainless steel improves its resistance to corrosion, which the set difference picks out immediately.