4 from numbers import Number
5 from typing import Generic, Iterable, List, Optional, Tuple, TypeVar
7 from math_utils import RunningMedian
8 from text_utils import bar_graph
11 T = TypeVar("T", bound=Number)
14 class SimpleHistogram(Generic[T]):
16 # Useful in defining wide open bottom/top bucket bounds:
17 POSITIVE_INFINITY = math.inf
18 NEGATIVE_INFINITY = -math.inf
20 def __init__(self, buckets: List[Tuple[T, T]]):
22 for start_end in buckets:
23 if self._get_bucket(start_end[0]) is not None:
24 raise Exception("Buckets overlap?!")
25 self.buckets[start_end] = 0
27 self.median = RunningMedian()
33 def n_evenly_spaced_buckets(
37 ) -> List[Tuple[T, T]]:
39 stride = int((max_bound - min_bound) / n)
41 raise Exception("Min must be < Max")
42 for bucket_start in range(min_bound, max_bound, stride):
43 ret.append((bucket_start, bucket_start + stride))
46 def _get_bucket(self, item: T) -> Optional[Tuple[T, T]]:
47 for start_end in self.buckets:
48 if start_end[0] <= item < start_end[1]:
52 def add_item(self, item: T) -> bool:
53 bucket = self._get_bucket(item)
57 self.buckets[bucket] += 1
59 self.median.add_number(item)
60 if self.maximum is None or item > self.maximum:
62 if self.minimum is None or item < self.minimum:
66 def add_items(self, lst: Iterable[T]) -> bool:
69 all_true = all_true and self.add_item(item)
72 def __repr__(self) -> str:
73 max_population: Optional[int] = None
74 for bucket in self.buckets:
75 pop = self.buckets[bucket]
77 last_bucket_start = bucket[0]
78 if max_population is None or pop > max_population:
81 if max_population is None:
84 for bucket in sorted(self.buckets, key=lambda x : x[0]):
85 pop = self.buckets[bucket]
89 (pop / max_population),
94 label = f'{start}..{end}'
95 txt += f'{label:12}: ' + bar + f"({pop}) ({len(bar)})\n"
96 if start == last_bucket_start:
99 txt = txt + f'''{self.count} item(s)
102 {self.sigma/self.count:.3f} mean
103 {self.median.get_median()} median'''