Accumulate #
The function accumulate takes s : α → Set β with LE α and returns ⋃ y ≤ x, s y.
It is related to dissipate s := ⋂ y ≤ x, s y.
accumulate is closely related to the function partialSups, although these two functions have
slightly different typeclass assumptions and API. partialSups_eq_accumulate shows
that they coincide on ℕ.
accumulate s is the union of s y for y ≤ x.
Equations
- Set.accumulate s x = ⋃ (y : α), ⋃ (_ : y ≤ x), s y
Instances For
@[simp]
theorem Set.subset_accumulate {α : Type u_1} {β : Type u_2} {s : α → Set β} [Preorder α] {x : α} :
s x ⊆ accumulate s x
theorem Set.accumulate_subset_iUnion {α : Type u_1} {β : Type u_2} {s : α → Set β} [LE α] (x : α) :
accumulate s x ⊆ ⋃ (i : α), s i
theorem Set.monotone_accumulate {α : Type u_1} {β : Type u_2} {s : α → Set β} [Preorder α] :
Monotone (accumulate s)
theorem Set.accumulate_subset_accumulate {α : Type u_1} {β : Type u_2} {s : α → Set β} [Preorder α] {x y : α} (h : x ≤ y) :
accumulate s x ⊆ accumulate s y
@[simp]
@[simp]
@[simp]
theorem Set.accumulate_bot {α : Type u_1} {β : Type u_2} [PartialOrder α] [OrderBot α] (s : α → Set β) :
theorem Set.disjoint_accumulate {α : Type u_1} {β : Type u_2} {s : α → Set β} [Preorder α] (hs : Pairwise (Function.onFun Disjoint s)) {i j : α} (hij : i < j) :
Disjoint (accumulate s i) (s j)
@[simp]
theorem Set.exists_subset_accumulate_of_directed {α : Type u_1} {s : ℕ → Set α} (hd : Directed (fun (x1 x2 : Set α) => x1 ⊆ x2) s) (n : ℕ) :
∃ (m : ℕ), accumulate s n ⊆ s m
For a directed set of sets s : ℕ → Set α and n : ℕ, there exists m : ℕ (maybe
larger than n) such that accumulate s n ⊆ s m.
theorem Set.directed_accumulate {α : Type u_1} {s : ℕ → Set α} :
Directed (fun (x1 x2 : Set α) => x1 ⊆ x2) (accumulate s)