Built-in types
int, float, str, bool, list, tuple, dict, set, None — the nine workhorses.
Code example
counts = [3, 1, 4, 1, 5] # list — mutable
point = (3.0, 4.0) # tuple — immutable
config = {'lr': 0.01, 'n': 100} # dict
flags = {'fit', 'predict'} # set
nothing = None # null sentinel
How it works
Python ships with nine built-in types that cover 95% of everyday code. Lists are ordered and mutable; tuples are ordered and immutable; dicts are key→value mappings (insertion-ordered since Python 3.7); sets are unordered unique collections. The benchmark pipeline uses lists for prediction arrays, dicts for metric configs, and tuples for fixed shapes like (n_samples, n_features).
🔍 Where this is used in OpenBenchML
app/routes/benchmark.py uses dicts for metric configs and lists for prediction arrays.