fact
stringlengths
6
3.84k
type
stringclasses
11 values
library
stringclasses
32 values
imports
listlengths
1
14
filename
stringlengths
20
95
symbolic_name
stringlengths
1
90
docstring
stringlengths
7
20k
id₁? (e : Expr) : MonoidalM (Option Obj) := do let ctx ← read match ctx.instMonoidal? with | some _monoidal => do if ← withDefault <| isDefEq e (q(𝟙_ _) : Q($ctx.C)) then return some ⟨none⟩ else return none | _ => return none
def
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Datatypes", "Mathlib.Tactic.CategoryTheory.MonoidalComp" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Datatypes.lean
id₁
Check that `e` is definitionally equal to `𝟙_ C`.
comp? (e : Expr) : MonoidalM (Option (Mor₁ × Mor₁)) := do let ctx ← read let f ← mkFreshExprMVarQ ctx.C let g ← mkFreshExprMVarQ ctx.C match ctx.instMonoidal? with | some _monoidal => do if ← withDefault <| isDefEq e q($f ⊗ $g) then let f ← instantiateMVars f let g ← instantiateMVars g return some ((.of ⟨f, ⟨none⟩, ⟨none⟩⟩ : Mor₁), (.of ⟨g, ⟨none⟩, ⟨none⟩⟩ : Mor₁)) else return none | _ => return none
def
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Datatypes", "Mathlib.Tactic.CategoryTheory.MonoidalComp" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Datatypes.lean
comp
Return `(f, g)` if `e` is definitionally equal to `f ⊗ g`.
partial mor₁OfExpr (e : Expr) : MonoidalM Mor₁ := do if let some f := (← get).cache.find? e then return f let f ← if let some a ← id₁? e then MonadMor₁.id₁M a else if let some (f, g) ← comp? e then MonadMor₁.comp₁M (← mor₁OfExpr f.e) (← mor₁OfExpr g.e) else return Mor₁.of ⟨e, ⟨none⟩, ⟨none⟩⟩ modify fun s => { s with cache := s.cache.insert e f } return f
def
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Datatypes", "Mathlib.Tactic.CategoryTheory.MonoidalComp" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Datatypes.lean
mor₁OfExpr
Construct a `Mor₁` expression from a Lean expression.
partial Mor₂IsoOfExpr (e : Expr) : MonoidalM Mor₂Iso := do match (← whnfR e).getAppFnArgs with | (``MonoidalCategoryStruct.associator, #[_, _, _, f, g, h]) => associatorM' (← MkMor₁.ofExpr f) (← MkMor₁.ofExpr g) (← MkMor₁.ofExpr h) | (``MonoidalCategoryStruct.leftUnitor, #[_, _, _, f]) => leftUnitorM' (← MkMor₁.ofExpr f) | (``MonoidalCategoryStruct.rightUnitor, #[_, _, _, f]) => rightUnitorM' (← MkMor₁.ofExpr f) | (``Iso.refl, #[_, _, f]) => id₂M' (← MkMor₁.ofExpr f) | (``Iso.symm, #[_, _, _, _, η]) => symmM (← Mor₂IsoOfExpr η) | (``Iso.trans, #[_, _, _, _, _, η, θ]) => comp₂M (← Mor₂IsoOfExpr η) (← Mor₂IsoOfExpr θ) | (``MonoidalCategory.whiskerLeftIso, #[_, _, _, f, _, _, η]) => whiskerLeftM (← MkMor₁.ofExpr f) (← Mor₂IsoOfExpr η) | (``MonoidalCategory.whiskerRightIso, #[_, _, _, _, _, η, h]) => whiskerRightM (← Mor₂IsoOfExpr η) (← MkMor₁.ofExpr h) | (``tensorIso, #[_, _, _, _, _, _, _, η, θ]) => horizontalCompM (← Mor₂IsoOfExpr η) (← Mor₂IsoOfExpr θ) | (``monoidalIsoComp, #[_, _, _, g, h, _, inst, η, θ]) => let α ← coherenceHomM (← MkMor₁.ofExpr g) (← MkMor₁.ofExpr h) inst coherenceCompM α (← Mor₂IsoOfExpr η) (← Mor₂IsoOfExpr θ) | (``MonoidalCoherence.iso, #[_, _, f, g, inst]) => coherenceHomM' (← MkMor₁.ofExpr f) (← MkMor₁.ofExpr g) inst | _ => return .of ⟨e, ← MkMor₁.ofExpr (← srcExprOfIso e), ← MkMor₁.ofExpr (← tgtExprOfIso e)⟩ open MonadMor₂ in
def
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Datatypes", "Mathlib.Tactic.CategoryTheory.MonoidalComp" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Datatypes.lean
Mor₂IsoOfExpr
Construct a `Mor₂Iso` term from a Lean expression.
partial Mor₂OfExpr (e : Expr) : MonoidalM Mor₂ := do match ← whnfR e with | .proj ``Iso 0 η => homM (← Mor₂IsoOfExpr η) | .proj ``Iso 1 η => invM (← Mor₂IsoOfExpr η) | .app .. => match (← whnfR e).getAppFnArgs with | (``CategoryStruct.id, #[_, _, f]) => id₂M (← MkMor₁.ofExpr f) | (``CategoryStruct.comp, #[_, _, _, _, _, η, θ]) => comp₂M (← Mor₂OfExpr η) (← Mor₂OfExpr θ) | (``MonoidalCategoryStruct.whiskerLeft, #[_, _, _, f, _, _, η]) => whiskerLeftM (← MkMor₁.ofExpr f) (← Mor₂OfExpr η) | (``MonoidalCategoryStruct.whiskerRight, #[_, _, _, _, _, η, h]) => whiskerRightM (← Mor₂OfExpr η) (← MkMor₁.ofExpr h) | (``MonoidalCategoryStruct.tensorHom, #[_, _, _, _, _, _, _, η, θ]) => horizontalCompM (← Mor₂OfExpr η) (← Mor₂OfExpr θ) | (``monoidalComp, #[_, _, _, g, h, _, inst, η, θ]) => let α ← coherenceHomM (← MkMor₁.ofExpr g) (← MkMor₁.ofExpr h) inst coherenceCompM α (← Mor₂OfExpr η) (← Mor₂OfExpr θ) | _ => return .of ⟨e, ← MkMor₁.ofExpr (← srcExpr e), ← MkMor₁.ofExpr (← tgtExpr e)⟩ | _ => return .of ⟨e, ← MkMor₁.ofExpr (← srcExpr e), ← MkMor₁.ofExpr (← tgtExpr e)⟩
def
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Datatypes", "Mathlib.Tactic.CategoryTheory.MonoidalComp" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Datatypes.lean
Mor₂OfExpr
Construct a `Mor₂` term from a Lean expression.
@[nolint synTaut] evalComp_nil_nil {f g h : C} (α : f ≅ g) (β : g ≅ h) : (α ≪≫ β).hom = (α ≪≫ β).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalComp_nil_nil
null
evalComp_nil_cons {f g h i j : C} (α : f ≅ g) (β : g ≅ h) (η : h ⟶ i) (ηs : i ⟶ j) : α.hom ≫ (β.hom ≫ η ≫ ηs) = (α ≪≫ β).hom ≫ η ≫ ηs := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalComp_nil_cons
null
evalComp_cons {f g h i j : C} (α : f ≅ g) (η : g ⟶ h) {ηs : h ⟶ i} {θ : i ⟶ j} {ι : h ⟶ j} (e_ι : ηs ≫ θ = ι) : (α.hom ≫ η ≫ ηs) ≫ θ = α.hom ≫ η ≫ ι := by simp [e_ι]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalComp_cons
null
eval_comp {η η' : f ⟶ g} {θ θ' : g ⟶ h} {ι : f ⟶ h} (e_η : η = η') (e_θ : θ = θ') (e_ηθ : η' ≫ θ' = ι) : η ≫ θ = ι := by simp [e_η, e_θ, e_ηθ]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
eval_comp
null
eval_of (η : f ⟶ g) : η = (Iso.refl _).hom ≫ η ≫ (Iso.refl _).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
eval_of
null
eval_monoidalComp {η η' : f ⟶ g} {α : g ≅ h} {θ θ' : h ⟶ i} {αθ : g ⟶ i} {ηαθ : f ⟶ i} (e_η : η = η') (e_θ : θ = θ') (e_αθ : α.hom ≫ θ' = αθ) (e_ηαθ : η' ≫ αθ = ηαθ) : η ≫ α.hom ≫ θ = ηαθ := by simp [e_η, e_θ, e_αθ, e_ηαθ] variable [MonoidalCategory C] @[nolint synTaut]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
eval_monoidalComp
null
evalWhiskerLeft_nil (f : C) {g h : C} (α : g ≅ h) : (whiskerLeftIso f α).hom = (whiskerLeftIso f α).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerLeft_nil
null
evalWhiskerLeft_of_cons {f g h i j : C} (α : g ≅ h) (η : h ⟶ i) {ηs : i ⟶ j} {θ : f ⊗ i ⟶ f ⊗ j} (e_θ : f ◁ ηs = θ) : f ◁ (α.hom ≫ η ≫ ηs) = (whiskerLeftIso f α).hom ≫ f ◁ η ≫ θ := by simp [e_θ]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerLeft_of_cons
null
evalWhiskerLeft_comp {f g h i : C} {η : h ⟶ i} {η₁ : g ⊗ h ⟶ g ⊗ i} {η₂ : f ⊗ g ⊗ h ⟶ f ⊗ g ⊗ i} {η₃ : f ⊗ g ⊗ h ⟶ (f ⊗ g) ⊗ i} {η₄ : (f ⊗ g) ⊗ h ⟶ (f ⊗ g) ⊗ i} (e_η₁ : g ◁ η = η₁) (e_η₂ : f ◁ η₁ = η₂) (e_η₃ : η₂ ≫ (α_ _ _ _).inv = η₃) (e_η₄ : (α_ _ _ _).hom ≫ η₃ = η₄) : (f ⊗ g) ◁ η = η₄ := by simp [e_η₁, e_η₂, e_η₃, e_η₄]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerLeft_comp
null
evalWhiskerLeft_id {f g : C} {η : f ⟶ g} {η₁ : f ⟶ 𝟙_ C ⊗ g} {η₂ : 𝟙_ C ⊗ f ⟶ 𝟙_ C ⊗ g} (e_η₁ : η ≫ (λ_ _).inv = η₁) (e_η₂ : (λ_ _).hom ≫ η₁ = η₂) : 𝟙_ C ◁ η = η₂ := by simp [e_η₁, e_η₂]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerLeft_id
null
eval_whiskerLeft {f g h : C} {η η' : g ⟶ h} {θ : f ⊗ g ⟶ f ⊗ h} (e_η : η = η') (e_θ : f ◁ η' = θ) : f ◁ η = θ := by simp [e_η, e_θ]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
eval_whiskerLeft
null
eval_whiskerRight {f g h : C} {η η' : f ⟶ g} {θ : f ⊗ h ⟶ g ⊗ h} (e_η : η = η') (e_θ : η' ▷ h = θ) : η ▷ h = θ := by simp [e_η, e_θ]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
eval_whiskerRight
null
eval_tensorHom {f g h i : C} {η η' : f ⟶ g} {θ θ' : h ⟶ i} {ι : f ⊗ h ⟶ g ⊗ i} (e_η : η = η') (e_θ : θ = θ') (e_ι : η' ⊗ₘ θ' = ι) : η ⊗ₘ θ = ι := by simp [e_η, e_θ, e_ι] @[nolint synTaut]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
eval_tensorHom
null
evalWhiskerRight_nil {f g : C} (α : f ≅ g) (h : C) : (whiskerRightIso α h).hom = (whiskerRightIso α h).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRight_nil
null
evalWhiskerRight_cons_of_of {f g h i j : C} {α : f ≅ g} {η : g ⟶ h} {ηs : h ⟶ i} {ηs₁ : h ⊗ j ⟶ i ⊗ j} {η₁ : g ⊗ j ⟶ h ⊗ j} {η₂ : g ⊗ j ⟶ i ⊗ j} {η₃ : f ⊗ j ⟶ i ⊗ j} (e_ηs₁ : ηs ▷ j = ηs₁) (e_η₁ : η ▷ j = η₁) (e_η₂ : η₁ ≫ ηs₁ = η₂) (e_η₃ : (whiskerRightIso α j).hom ≫ η₂ = η₃) : (α.hom ≫ η ≫ ηs) ▷ j = η₃ := by simp_all
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRight_cons_of_of
null
evalWhiskerRight_cons_whisker {f g h i j k : C} {α : g ≅ f ⊗ h} {η : h ⟶ i} {ηs : f ⊗ i ⟶ j} {η₁ : h ⊗ k ⟶ i ⊗ k} {η₂ : f ⊗ (h ⊗ k) ⟶ f ⊗ (i ⊗ k)} {ηs₁ : (f ⊗ i) ⊗ k ⟶ j ⊗ k} {ηs₂ : f ⊗ (i ⊗ k) ⟶ j ⊗ k} {η₃ : f ⊗ (h ⊗ k) ⟶ j ⊗ k} {η₄ : (f ⊗ h) ⊗ k ⟶ j ⊗ k} {η₅ : g ⊗ k ⟶ j ⊗ k} (e_η₁ : ((Iso.refl _).hom ≫ η ≫ (Iso.refl _).hom) ▷ k = η₁) (e_η₂ : f ◁ η₁ = η₂) (e_ηs₁ : ηs ▷ k = ηs₁) (e_ηs₂ : (α_ _ _ _).inv ≫ ηs₁ = ηs₂) (e_η₃ : η₂ ≫ ηs₂ = η₃) (e_η₄ : (α_ _ _ _).hom ≫ η₃ = η₄) (e_η₅ : (whiskerRightIso α k).hom ≫ η₄ = η₅) : (α.hom ≫ (f ◁ η) ≫ ηs) ▷ k = η₅ := by simp at e_η₁ e_η₅ simp [e_η₁, e_η₂, e_ηs₁, e_ηs₂, e_η₃, e_η₄, e_η₅]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRight_cons_whisker
null
evalWhiskerRight_comp {f f' g h : C} {η : f ⟶ f'} {η₁ : f ⊗ g ⟶ f' ⊗ g} {η₂ : (f ⊗ g) ⊗ h ⟶ (f' ⊗ g) ⊗ h} {η₃ : (f ⊗ g) ⊗ h ⟶ f' ⊗ (g ⊗ h)} {η₄ : f ⊗ (g ⊗ h) ⟶ f' ⊗ (g ⊗ h)} (e_η₁ : η ▷ g = η₁) (e_η₂ : η₁ ▷ h = η₂) (e_η₃ : η₂ ≫ (α_ _ _ _).hom = η₃) (e_η₄ : (α_ _ _ _).inv ≫ η₃ = η₄) : η ▷ (g ⊗ h) = η₄ := by simp [e_η₁, e_η₂, e_η₃, e_η₄]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRight_comp
null
evalWhiskerRight_id {f g : C} {η : f ⟶ g} {η₁ : f ⟶ g ⊗ 𝟙_ C} {η₂ : f ⊗ 𝟙_ C ⟶ g ⊗ 𝟙_ C} (e_η₁ : η ≫ (ρ_ _).inv = η₁) (e_η₂ : (ρ_ _).hom ≫ η₁ = η₂) : η ▷ 𝟙_ C = η₂ := by simp [e_η₁, e_η₂]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRight_id
null
evalWhiskerRightAux_of {f g : C} (η : f ⟶ g) (h : C) : η ▷ h = (Iso.refl _).hom ≫ η ▷ h ≫ (Iso.refl _).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRightAux_of
null
evalWhiskerRightAux_cons {f g h i j : C} {η : g ⟶ h} {ηs : i ⟶ j} {ηs' : i ⊗ f ⟶ j ⊗ f} {η₁ : g ⊗ (i ⊗ f) ⟶ h ⊗ (j ⊗ f)} {η₂ : g ⊗ (i ⊗ f) ⟶ (h ⊗ j) ⊗ f} {η₃ : (g ⊗ i) ⊗ f ⟶ (h ⊗ j) ⊗ f} (e_ηs' : ηs ▷ f = ηs') (e_η₁ : ((Iso.refl _).hom ≫ η ≫ (Iso.refl _).hom) ⊗ₘ ηs' = η₁) (e_η₂ : η₁ ≫ (α_ _ _ _).inv = η₂) (e_η₃ : (α_ _ _ _).hom ≫ η₂ = η₃) : (η ⊗ₘ ηs) ▷ f = η₃ := by simp [← e_ηs', ← e_η₁, ← e_η₂, ← e_η₃, MonoidalCategory.tensorHom_def]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRightAux_cons
null
evalWhiskerRight_cons_of {f f' g h i : C} {α : f' ≅ g} {η : g ⟶ h} {ηs : h ⟶ i} {ηs₁ : h ⊗ f ⟶ i ⊗ f} {η₁ : g ⊗ f ⟶ h ⊗ f} {η₂ : g ⊗ f ⟶ i ⊗ f} {η₃ : f' ⊗ f ⟶ i ⊗ f} (e_ηs₁ : ηs ▷ f = ηs₁) (e_η₁ : η ▷ f = η₁) (e_η₂ : η₁ ≫ ηs₁ = η₂) (e_η₃ : (whiskerRightIso α f).hom ≫ η₂ = η₃) : (α.hom ≫ η ≫ ηs) ▷ f = η₃ := by simp_all
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalWhiskerRight_cons_of
null
evalHorizontalCompAux_of {f g h i : C} (η : f ⟶ g) (θ : h ⟶ i) : η ⊗ₘ θ = (Iso.refl _).hom ≫ (η ⊗ₘ θ) ≫ (Iso.refl _).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalCompAux_of
null
evalHorizontalCompAux_cons {f f' g g' h i : C} {η : f ⟶ g} {ηs : f' ⟶ g'} {θ : h ⟶ i} {ηθ : f' ⊗ h ⟶ g' ⊗ i} {η₁ : f ⊗ (f' ⊗ h) ⟶ g ⊗ (g' ⊗ i)} {ηθ₁ : f ⊗ (f' ⊗ h) ⟶ (g ⊗ g') ⊗ i} {ηθ₂ : (f ⊗ f') ⊗ h ⟶ (g ⊗ g') ⊗ i} (e_ηθ : ηs ⊗ₘ θ = ηθ) (e_η₁ : ((Iso.refl _).hom ≫ η ≫ (Iso.refl _).hom) ⊗ₘ ηθ = η₁) (e_ηθ₁ : η₁ ≫ (α_ _ _ _).inv = ηθ₁) (e_ηθ₂ : (α_ _ _ _).hom ≫ ηθ₁ = ηθ₂) : (η ⊗ₘ ηs) ⊗ₘ θ = ηθ₂ := by simp_all
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalCompAux_cons
null
evalHorizontalCompAux'_whisker {f f' g g' h : C} {η : g ⟶ h} {θ : f' ⟶ g'} {ηθ : g ⊗ f' ⟶ h ⊗ g'} {η₁ : f ⊗ (g ⊗ f') ⟶ f ⊗ (h ⊗ g')} {η₂ : f ⊗ (g ⊗ f') ⟶ (f ⊗ h) ⊗ g'} {η₃ : (f ⊗ g) ⊗ f' ⟶ (f ⊗ h) ⊗ g'} (e_ηθ : η ⊗ₘ θ = ηθ) (e_η₁ : f ◁ ηθ = η₁) (e_η₂ : η₁ ≫ (α_ _ _ _).inv = η₂) (e_η₃ : (α_ _ _ _).hom ≫ η₂ = η₃) : (f ◁ η) ⊗ₘ θ = η₃ := by simp only [← e_ηθ, ← e_η₁, ← e_η₂, ← e_η₃] simp [MonoidalCategory.tensorHom_def]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalCompAux'_whisker
null
evalHorizontalCompAux'_of_whisker {f f' g g' h : C} {η : g ⟶ h} {θ : f' ⟶ g'} {η₁ : g ⊗ f ⟶ h ⊗ f} {ηθ : (g ⊗ f) ⊗ f' ⟶ (h ⊗ f) ⊗ g'} {ηθ₁ : (g ⊗ f) ⊗ f' ⟶ h ⊗ (f ⊗ g')} {ηθ₂ : g ⊗ (f ⊗ f') ⟶ h ⊗ (f ⊗ g')} (e_η₁ : η ▷ f = η₁) (e_ηθ : η₁ ⊗ₘ ((Iso.refl _).hom ≫ θ ≫ (Iso.refl _).hom) = ηθ) (e_ηθ₁ : ηθ ≫ (α_ _ _ _).hom = ηθ₁) (e_ηθ₂ : (α_ _ _ _).inv ≫ ηθ₁ = ηθ₂) : η ⊗ₘ (f ◁ θ) = ηθ₂ := by simp only [← e_η₁, ← e_ηθ, ← e_ηθ₁, ← e_ηθ₂] simp [MonoidalCategory.tensorHom_def] @[nolint synTaut]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalCompAux'_of_whisker
null
evalHorizontalComp_nil_nil {f g h i : C} (α : f ≅ g) (β : h ≅ i) : (α ⊗ᵢ β).hom = (α ⊗ᵢ β).hom := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalComp_nil_nil
null
evalHorizontalComp_nil_cons {f f' g g' h i : C} {α : f ≅ g} {β : f' ≅ g'} {η : g' ⟶ h} {ηs : h ⟶ i} {η₁ : g ⊗ g' ⟶ g ⊗ h} {ηs₁ : g ⊗ h ⟶ g ⊗ i} {η₂ : g ⊗ g' ⟶ g ⊗ i} {η₃ : f ⊗ f' ⟶ g ⊗ i} (e_η₁ : g ◁ ((Iso.refl _).hom ≫ η ≫ (Iso.refl _).hom) = η₁) (e_ηs₁ : g ◁ ηs = ηs₁) (e_η₂ : η₁ ≫ ηs₁ = η₂) (e_η₃ : (α ⊗ᵢ β).hom ≫ η₂ = η₃) : α.hom ⊗ₘ (β.hom ≫ η ≫ ηs) = η₃ := by simp_all [MonoidalCategory.tensorHom_def]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalComp_nil_cons
null
evalHorizontalComp_cons_nil {f f' g g' h i : C} {α : f ≅ g} {η : g ⟶ h} {ηs : h ⟶ i} {β : f' ≅ g'} {η₁ : g ⊗ g' ⟶ h ⊗ g'} {ηs₁ : h ⊗ g' ⟶ i ⊗ g'} {η₂ : g ⊗ g' ⟶ i ⊗ g'} {η₃ : f ⊗ f' ⟶ i ⊗ g'} (e_η₁ : ((Iso.refl _).hom ≫ η ≫ (Iso.refl _).hom) ▷ g' = η₁) (e_ηs₁ : ηs ▷ g' = ηs₁) (e_η₂ : η₁ ≫ ηs₁ = η₂) (e_η₃ : (α ⊗ᵢ β).hom ≫ η₂ = η₃) : (α.hom ≫ η ≫ ηs) ⊗ₘ β.hom = η₃ := by simp_all [MonoidalCategory.tensorHom_def']
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalComp_cons_nil
null
evalHorizontalComp_cons_cons {f f' g g' h h' i i' : C} {α : f ≅ g} {η : g ⟶ h} {ηs : h ⟶ i} {β : f' ≅ g'} {θ : g' ⟶ h'} {θs : h' ⟶ i'} {ηθ : g ⊗ g' ⟶ h ⊗ h'} {ηθs : h ⊗ h' ⟶ i ⊗ i'} {ηθ₁ : g ⊗ g' ⟶ i ⊗ i'} {ηθ₂ : f ⊗ f' ⟶ i ⊗ i'} (e_ηθ : η ⊗ₘ θ = ηθ) (e_ηθs : ηs ⊗ₘ θs = ηθs) (e_ηθ₁ : ηθ ≫ ηθs = ηθ₁) (e_ηθ₂ : (α ⊗ᵢ β).hom ≫ ηθ₁ = ηθ₂) : (α.hom ≫ η ≫ ηs) ⊗ₘ (β.hom ≫ θ ≫ θs) = ηθ₂ := by simp [← e_ηθ, ← e_ηθs, ← e_ηθ₁, ← e_ηθ₂]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.Normalize", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/Normalize.lean
evalHorizontalComp_cons_cons
null
normalizeIsoComp {p f g pf pfg : C} (η_f : p ⊗ f ≅ pf) (η_g : pf ⊗ g ≅ pfg) := (α_ _ _ _).symm ≪≫ whiskerRightIso η_f g ≪≫ η_g
abbrev
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
normalizeIsoComp
The composition of the normalizing isomorphisms `η_f : p ⊗ f ≅ pf` and `η_g : pf ⊗ g ≅ pfg`.
naturality_associator {p f g h pf pfg pfgh : C} (η_f : p ⊗ f ≅ pf) (η_g : pf ⊗ g ≅ pfg) (η_h : pfg ⊗ h ≅ pfgh) : p ◁ (α_ f g h) ≪≫ normalizeIsoComp η_f (normalizeIsoComp η_g η_h) = normalizeIsoComp (normalizeIsoComp η_f η_g) η_h := Iso.ext (by simp)
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_associator
null
naturality_leftUnitor {p f pf : C} (η_f : p ⊗ f ≅ pf) : p ◁ (λ_ f) ≪≫ η_f = normalizeIsoComp (ρ_ p) η_f := Iso.ext (by simp)
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_leftUnitor
null
naturality_rightUnitor {p f pf : C} (η_f : p ⊗ f ≅ pf) : p ◁ (ρ_ f) ≪≫ η_f = normalizeIsoComp η_f (ρ_ pf) := Iso.ext (by simp)
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_rightUnitor
null
naturality_id {p f pf : C} (η_f : p ⊗ f ≅ pf) : p ◁ Iso.refl f ≪≫ η_f = η_f := by simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_id
null
naturality_comp {p f g h pf : C} {η : f ≅ g} {θ : g ≅ h} (η_f : p ⊗ f ≅ pf) (η_g : p ⊗ g ≅ pf) (η_h : p ⊗ h ≅ pf) (ih_η : p ◁ η ≪≫ η_g = η_f) (ih_θ : p ◁ θ ≪≫ η_h = η_g) : p ◁ (η ≪≫ θ) ≪≫ η_h = η_f := by simp_all
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_comp
null
naturality_whiskerLeft {p f g h pf pfg : C} {η : g ≅ h} (η_f : p ⊗ f ≅ pf) (η_fg : pf ⊗ g ≅ pfg) (η_fh : (pf ⊗ h) ≅ pfg) (ih_η : pf ◁ η ≪≫ η_fh = η_fg) : p ◁ (f ◁ η) ≪≫ normalizeIsoComp η_f η_fh = normalizeIsoComp η_f η_fg := by rw [← ih_η] apply Iso.ext simp [← whisker_exchange_assoc]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_whiskerLeft
null
naturality_whiskerRight {p f g h pf pfh : C} {η : f ≅ g} (η_f : p ⊗ f ≅ pf) (η_g : p ⊗ g ≅ pf) (η_fh : (pf ⊗ h) ≅ pfh) (ih_η : p ◁ η ≪≫ η_g = η_f) : p ◁ (η ▷ h) ≪≫ normalizeIsoComp η_g η_fh = normalizeIsoComp η_f η_fh := by rw [← ih_η] apply Iso.ext simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_whiskerRight
null
naturality_tensorHom {p f₁ g₁ f₂ g₂ pf₁ pf₁f₂ : C} {η : f₁ ≅ g₁} {θ : f₂ ≅ g₂} (η_f₁ : p ⊗ f₁ ≅ pf₁) (η_g₁ : p ⊗ g₁ ≅ pf₁) (η_f₂ : pf₁ ⊗ f₂ ≅ pf₁f₂) (η_g₂ : pf₁ ⊗ g₂ ≅ pf₁f₂) (ih_η : p ◁ η ≪≫ η_g₁ = η_f₁) (ih_θ : pf₁ ◁ θ ≪≫ η_g₂ = η_f₂) : p ◁ (η ⊗ᵢ θ) ≪≫ normalizeIsoComp η_g₁ η_g₂ = normalizeIsoComp η_f₁ η_f₂ := by rw [tensorIso_def] apply naturality_comp · apply naturality_whiskerRight _ _ _ ih_η · apply naturality_whiskerLeft _ _ _ ih_θ
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_tensorHom
null
naturality_inv {p f g pf : C} {η : f ≅ g} (η_f : p ⊗ f ≅ pf) (η_g : p ⊗ g ≅ pf) (ih : p ◁ η ≪≫ η_g = η_f) : p ◁ η.symm ≪≫ η_f = η_g := by rw [← ih] apply Iso.ext simp
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
naturality_inv
null
of_normalize_eq {f g f' : C} {η θ : f ≅ g} (η_f : 𝟙_ C ⊗ f ≅ f') (η_g : 𝟙_ C ⊗ g ≅ f') (h_η : 𝟙_ C ◁ η ≪≫ η_g = η_f) (h_θ : 𝟙_ C ◁ θ ≪≫ η_g = η_f) : η = θ := by apply Iso.ext calc η.hom = (λ_ f).inv ≫ η_f.hom ≫ η_g.inv ≫ (λ_ g).hom := by simp [← reassoc_of% (congrArg Iso.hom h_η)] _ = θ.hom := by simp [← reassoc_of% (congrArg Iso.hom h_θ)]
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
of_normalize_eq
null
mk_eq_of_naturality {f g f' : C} {η θ : f ⟶ g} {η' θ' : f ≅ g} (η_f : 𝟙_ C ⊗ f ≅ f') (η_g : 𝟙_ C ⊗ g ≅ f') (η_hom : η'.hom = η) (Θ_hom : θ'.hom = θ) (Hη : whiskerLeftIso (𝟙_ C) η' ≪≫ η_g = η_f) (Hθ : whiskerLeftIso (𝟙_ C) θ' ≪≫ η_g = η_f) : η = θ := calc η = η'.hom := η_hom.symm _ = (λ_ f).inv ≫ η_f.hom ≫ η_g.inv ≫ (λ_ g).hom := by simp [← reassoc_of% (congrArg Iso.hom Hη)] _ = θ'.hom := by simp [← reassoc_of% (congrArg Iso.hom Hθ)] _ = θ := Θ_hom
theorem
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
mk_eq_of_naturality
null
pureCoherence (mvarId : MVarId) : MetaM (List MVarId) := BicategoryLike.pureCoherence Monoidal.Context `monoidal mvarId @[inherit_doc pureCoherence] elab "monoidal_coherence" : tactic => withMainContext do replaceMainGoal <| ← Monoidal.pureCoherence <| ← getMainGoal
def
Tactic
[ "Mathlib.Tactic.CategoryTheory.Coherence.PureCoherence", "Mathlib.Tactic.CategoryTheory.Monoidal.Datatypes" ]
Mathlib/Tactic/CategoryTheory/Monoidal/PureCoherence.lean
pureCoherence
Close the goal of the form `η = θ`, where `η` and `θ` are 2-isomorphisms made up only of associators, unitors, and identities. ```lean example {C : Type} [Category C] [MonoidalCategory C] : (λ_ (𝟙_ C)).hom = (ρ_ (𝟙_ C)).hom := by monoidal_coherence ```
union (t₁ t₂ : TreeSet α cmp) : TreeSet α cmp := t₂.foldl .insert t₁
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
union
`O(n₂ * log (n₁ + n₂))`. Merges the maps `t₁` and `t₂`. If equal keys exist in both, the key from `t₂` is preferred.
sdiff (t₁ t₂ : TreeSet α cmp) : TreeSet α cmp := t₁.filter (!t₂.contains ·)
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
sdiff
`O(n₁ * (log n₁ + log n₂))`. Constructs the set of all elements of `t₁` that are not in `t₂`.
CompSource : Type | assump : Nat → CompSource | add : CompSource → CompSource → CompSource | scale : Nat → CompSource → CompSource deriving Inhabited
inductive
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
CompSource
`CompSource` tracks the source of a comparison. The atomic source of a comparison is an assumption, indexed by a natural number. Two comparisons can be added to produce a new comparison, and one comparison can be scaled by a natural number to produce a new comparison.
CompSource.flatten : CompSource → Std.HashMap Nat Nat | (CompSource.assump n) => (∅ : Std.HashMap Nat Nat).insert n 1 | (CompSource.add c1 c2) => (CompSource.flatten c1).mergeWith (fun _ b b' => b + b') (CompSource.flatten c2) | (CompSource.scale n c) => (CompSource.flatten c).map (fun _ v => v * n)
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
CompSource.flatten
Given a `CompSource` `cs`, `cs.flatten` maps an assumption index to the number of copies of that assumption that appear in the history of `cs`. For example, suppose `cs` is produced by scaling assumption 2 by 5, and adding to that the sum of assumptions 1 and 2. `cs.flatten` maps `1 ↦ 1, 2 ↦ 6`.
CompSource.toString : CompSource → String | (CompSource.assump e) => ToString.toString e | (CompSource.add c1 c2) => CompSource.toString c1 ++ " + " ++ CompSource.toString c2 | (CompSource.scale n c) => ToString.toString n ++ " * " ++ CompSource.toString c
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
CompSource.toString
Formats a `CompSource` for printing.
PComp : Type where /-- The comparison `Σ cᵢ*xᵢ R 0`. -/ c : Comp /-- We track how the comparison was constructed by adding and scaling previous comparisons, back to the original assumptions. -/ src : CompSource /-- The set of original assumptions which have been used in constructing this comparison. -/ history : TreeSet ℕ Ord.compare /-- The variables which have been *effectively eliminated*, i.e. by running the elimination algorithm on that variable. -/ effective : TreeSet ℕ Ord.compare /-- The variables which have been *implicitly eliminated*. These are variables that appear in the historical set, do not appear in `c` itself, and are not in `effective. -/ implicit : TreeSet ℕ Ord.compare /-- The union of all variables appearing in those original assumptions which appear in the `history` set. -/ vars : TreeSet ℕ Ord.compare
structure
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp
A `PComp` stores a linear comparison `Σ cᵢ*xᵢ R 0`, along with information about how this comparison was derived. The original expressions fed into `linarith` are each assigned a unique natural number label. The *historical set* `PComp.history` stores the labels of expressions that were used in deriving the current `PComp`. Variables are also indexed by natural numbers. The sets `PComp.effective`, `PComp.implicit`, and `PComp.vars` contain variable indices. * `PComp.vars` contains the variables that appear in any inequality in the historical set. * `PComp.effective` contains the variables that have been effectively eliminated from `PComp`. A variable `n` is said to be *effectively eliminated* in `p : PComp` if the elimination of `n` produced at least one of the ancestors of `p` (or `p` itself). * `PComp.implicit` contains the variables that have been implicitly eliminated from `PComp`. A variable `n` is said to be *implicitly eliminated* in `p` if it satisfies the following properties: - `n` appears in some inequality in the historical set (i.e. in `p.vars`). - `n` does not appear in `p.c.vars` (i.e. it has been eliminated). - `n` was not effectively eliminated. We track these sets in order to compute whether the history of a `PComp` is *minimal*. Checking this directly is expensive, but effective approximations can be defined in terms of these sets. During the variable elimination process, a `PComp` with non-minimal history can be discarded.
PComp.maybeMinimal (c : PComp) (elimedGE : ℕ) : Bool := c.history.size ≤ 1 + ((c.implicit.filter (· ≥ elimedGE)).union c.effective).size
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp.maybeMinimal
Any comparison whose history is not minimal is redundant, and need not be included in the new set of comparisons. `elimedGE : ℕ` is a natural number such that all variables with index ≥ `elimedGE` have been removed from the system. This test is an overapproximation to minimality. It gives necessary but not sufficient conditions. If the history of `c` is minimal, then `c.maybeMinimal` is true, but `c.maybeMinimal` may also be true for some `c` with non-minimal history. Thus, if `c.maybeMinimal` is false, `c` is known not to be minimal and must be redundant. See https://doi.org/10.1016/B978-0-444-88771-9.50019-2 (Theorem 13). The condition described there considers only implicitly eliminated variables that have been officially eliminated from the system. This is not the case for every implicitly eliminated variable. Consider eliminating `z` from `{x + y + z < 0, x - y - z < 0}`. The result is the set `{2*x < 0}`; `y` is implicitly but not officially eliminated. This implementation of Fourier-Motzkin elimination processes variables in decreasing order of indices. Immediately after a step that eliminates variable `k`, variable `k'` has been eliminated iff `k' ≥ k`. Thus we can compute the intersection of officially and implicitly eliminated variables by taking the set of implicitly eliminated variables with indices ≥ `elimedGE`.
PComp.cmp (p1 p2 : PComp) : Ordering := p1.c.cmp p2.c
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp.cmp
The `src : CompSource` field is ignored when comparing `PComp`s. Two `PComp`s proving the same comparison, with different sources, are considered equivalent.
PComp.scale (c : PComp) (n : ℕ) : PComp := { c with c := c.c.scale n, src := c.src.scale n }
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp.scale
`PComp.scale c n` scales the coefficients of `c` by `n` and notes this in the `CompSource`.
PComp.add (c1 c2 : PComp) (elimVar : ℕ) : PComp := let c := c1.c.add c2.c let src := c1.src.add c2.src let history := c1.history.union c2.history let vars := c1.vars.union c2.vars let effective := (c1.effective.union c2.effective).insert elimVar let implicit := (vars.sdiff (.ofList c.vars _)).sdiff effective ⟨c, src, history, effective, implicit, vars⟩
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp.add
`PComp.add c1 c2 elimVar` creates the result of summing the linear comparisons `c1` and `c2`, during the process of eliminating the variable `elimVar`. The computation assumes, but does not enforce, that `elimVar` appears in both `c1` and `c2` and does not appear in the sum. Computing the sum of the two comparisons is easy; the complicated details lie in tracking the additional fields of `PComp`. * The historical set `pcomp.history` of `c1 + c2` is the union of the two historical sets. * `vars` is the union of `c1.vars` and `c2.vars`. * The effectively eliminated variables of `c1 + c2` are the union of the two effective sets, with `elim_var` inserted. * The implicitly eliminated variables of `c1 + c2` are those that appear in `vars` but not `c.vars` or `effective`. (Note that the description of the implicitly eliminated variables of `c1 + c2` in the algorithm described in Section 6 of https://doi.org/10.1016/B978-0-444-88771-9.50019-2 seems to be wrong: that says it should be `(c1.implicit.union c2.implicit).sdiff explicit`. Since the implicitly eliminated sets start off empty for the assumption, this formula would leave them always empty.)
PComp.assump (c : Comp) (n : ℕ) : PComp where c := c src := CompSource.assump n history := {n} effective := .empty implicit := .empty vars := .ofList c.vars _
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp.assump
`PComp.assump c n` creates a `PComp` whose comparison is `c` and whose source is `CompSource.assump n`, that is, `c` is derived from the `n`th hypothesis. The history is the singleton set `{n}`. No variables have been eliminated (effectively or implicitly).
PCompSet := TreeSet PComp PComp.cmp /-! ### Elimination procedure -/
abbrev
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PCompSet
A collection of comparisons.
elimVar (c1 c2 : Comp) (a : ℕ) : Option (ℕ × ℕ) := let v1 := c1.coeffOf a let v2 := c2.coeffOf a if v1 * v2 < 0 then let vlcm := Nat.lcm v1.natAbs v2.natAbs some ⟨vlcm / v1.natAbs, vlcm / v2.natAbs⟩ else none
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
elimVar
If `c1` and `c2` both contain variable `a` with opposite coefficients, produces `v1` and `v2` such that `a` has been cancelled in `v1*c1 + v2*c2`.
pelimVar (p1 p2 : PComp) (a : ℕ) : Option PComp := do let (n1, n2) ← elimVar p1.c p2.c a return (p1.scale n1).add (p2.scale n2) a
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
pelimVar
`pelimVar p1 p2` calls `elimVar` on the `Comp` components of `p1` and `p2`. If this returns `v1` and `v2`, it creates a new `PComp` equal to `v1*p1 + v2*p2`, and tracks this in the `CompSource`.
PComp.isContr (p : PComp) : Bool := p.c.isContr
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
PComp.isContr
A `PComp` represents a contradiction if its `Comp` field represents a contradiction.
elimWithSet (a : ℕ) (p : PComp) (comps : PCompSet) : PCompSet := comps.foldl (fun s pc => match pelimVar p pc a with | some pc => if pc.maybeMinimal a then s.insert pc else s | none => s) TreeSet.empty
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
elimWithSet
`elimWithSet a p comps` collects the result of calling `pelimVar p p' a` for every `p' ∈ comps`.
LinarithData : Type where /-- The largest variable index that has not been (officially) eliminated. -/ maxVar : ℕ /-- The set of comparisons. -/ comps : PCompSet
structure
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
LinarithData
The state for the elimination monad. * `maxVar`: the largest variable index that has not been eliminated. * `comps`: a set of comparisons The elimination procedure proceeds by eliminating variable `v` from `comps` progressively in decreasing order.
LinarithM : Type → Type := StateT LinarithData (ExceptT PComp Lean.Core.CoreM)
abbrev
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
LinarithM
The linarith monad extends an exceptional monad with a `LinarithData` state. An exception produces a contradictory `PComp`.
getMaxVar : LinarithM ℕ := LinarithData.maxVar <$> get
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
getMaxVar
Returns the current max variable.
getPCompSet : LinarithM PCompSet := LinarithData.comps <$> get
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
getPCompSet
Return the current comparison set.
validate : LinarithM Unit := do match (← getPCompSet).toList.find? (fun p : PComp => p.isContr) with | none => return () | some c => throwThe _ c
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
validate
Throws an exception if a contradictory `PComp` is contained in the current state.
update (maxVar : ℕ) (comps : PCompSet) : LinarithM Unit := do StateT.set ⟨maxVar, comps⟩ validate
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
update
Updates the current state with a new max variable and comparisons, and calls `validate` to check for a contradiction.
splitSetByVarSign (a : ℕ) (comps : PCompSet) : PCompSet × PCompSet × PCompSet := comps.foldl (fun ⟨pos, neg, notPresent⟩ pc => let n := pc.c.coeffOf a if n > 0 then ⟨pos.insert pc, neg, notPresent⟩ else if n < 0 then ⟨pos, neg.insert pc, notPresent⟩ else ⟨pos, neg, notPresent.insert pc⟩) ⟨TreeSet.empty, TreeSet.empty, TreeSet.empty⟩
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
splitSetByVarSign
`splitSetByVarSign a comps` partitions the set `comps` into three parts. * `pos` contains the elements of `comps` in which `a` has a positive coefficient. * `neg` contains the elements of `comps` in which `a` has a negative coefficient. * `notPresent` contains the elements of `comps` in which `a` has coefficient 0. Returns `(pos, neg, notPresent)`.
elimVarM (a : ℕ) : LinarithM Unit := do let vs ← getMaxVar if (a ≤ vs) then Lean.Core.checkSystem decl_name%.toString let ⟨pos, neg, notPresent⟩ := splitSetByVarSign a (← getPCompSet) update (vs - 1) (← pos.foldlM (fun s p => do Lean.Core.checkSystem decl_name%.toString pure (s.union (elimWithSet a p neg))) notPresent) else pure ()
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
elimVarM
`elimVarM a` performs one round of Fourier-Motzkin elimination, eliminating the variable `a` from the `linarith` state.
elimAllVarsM : LinarithM Unit := do for i in (List.range ((← getMaxVar) + 1)).reverse do elimVarM i
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
elimAllVarsM
`elimAllVarsM` eliminates all variables from the linarith state, leaving it with a set of ground comparisons. If this succeeds without exception, the original `linarith` state is consistent.
mkLinarithData (hyps : List Comp) (maxVar : ℕ) : LinarithData := ⟨maxVar, .ofList (hyps.mapIdx fun n cmp => PComp.assump cmp n) _⟩
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
mkLinarithData
`mkLinarithData hyps vars` takes a list of hypotheses and the largest variable present in those hypotheses. It produces an initial state for the elimination monad.
CertificateOracle.fourierMotzkin : CertificateOracle where produceCertificate hyps maxVar := do let linarithData := mkLinarithData hyps maxVar let result ← (ExceptT.run (StateT.run (do validate; elimAllVarsM : LinarithM Unit) linarithData) :) match result with | (Except.ok _) => failure | (Except.error contr) => return contr.src.flatten
def
Tactic
[ "Batteries.Lean.HashMap", "Mathlib.Tactic.Linarith.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/FourierMotzkin.lean
CertificateOracle.fourierMotzkin
An oracle that uses Fourier-Motzkin elimination.
preprocess (matType : ℕ → ℕ → Type) [UsableInSimplexAlgorithm matType] (hyps : List Comp) (maxVar : ℕ) : matType (maxVar + 1) (hyps.length) × List Nat := let values : List (ℕ × ℕ × ℚ) := hyps.foldlIdx (init := []) fun idx cur comp => cur ++ comp.coeffs.map fun (var, c) => (var, idx, c) let strictIndexes := hyps.findIdxs (·.str == Ineq.lt) (ofValues values, strictIndexes)
def
Tactic
[ "Mathlib.Tactic.Linarith.Datatypes", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.PositiveVector" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm.lean
preprocess
Preprocess the goal to pass it to `Linarith.SimplexAlgorithm.findPositiveVector`.
postprocess (vec : Array ℚ) : Std.HashMap ℕ ℕ := let common_den : ℕ := vec.foldl (fun acc item => acc.lcm item.den) 1 let vecNat : Array ℕ := vec.map (fun x : ℚ => (x * common_den).floor.toNat) (∅ : Std.HashMap Nat Nat).insertMany <| vecNat.zipIdx.filterMap fun ⟨item, idx⟩ => if item != 0 then some (idx, item) else none
def
Tactic
[ "Mathlib.Tactic.Linarith.Datatypes", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.PositiveVector" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm.lean
postprocess
Extract the certificate from the `vec` found by `Linarith.SimplexAlgorithm.findPositiveVector`.
CertificateOracle.simplexAlgorithmSparse : CertificateOracle where produceCertificate hyps maxVar := do let (A, strictIndexes) := preprocess SparseMatrix hyps maxVar let vec ← findPositiveVector A strictIndexes return postprocess vec
def
Tactic
[ "Mathlib.Tactic.Linarith.Datatypes", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.PositiveVector" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm.lean
CertificateOracle.simplexAlgorithmSparse
An oracle that uses the Simplex Algorithm.
CertificateOracle.simplexAlgorithmDense : CertificateOracle where produceCertificate hyps maxVar := do let (A, strictIndexes) := preprocess DenseMatrix hyps maxVar let vec ← findPositiveVector A strictIndexes return postprocess vec
def
Tactic
[ "Mathlib.Tactic.Linarith.Datatypes", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.PositiveVector" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm.lean
CertificateOracle.simplexAlgorithmDense
The same oracle as `CertificateOracle.simplexAlgorithmSparse`, but uses dense matrices. Works faster on dense states.
UsableInSimplexAlgorithm (α : Nat → Nat → Type) where /-- Returns `mat[i, j]`. -/ getElem {n m : Nat} (mat : α n m) (i j : Nat) : Rat /-- Sets `mat[i, j]`. -/ setElem {n m : Nat} (mat : α n m) (i j : Nat) (v : Rat) : α n m /-- Returns the list of elements of `mat` in the form `(i, j, mat[i, j])`. -/ getValues {n m : Nat} (mat : α n m) : List (Nat × Nat × Rat) /-- Creates a matrix from a list of elements in the form `(i, j, mat[i, j])`. -/ ofValues {n m : Nat} (values : List (Nat × Nat × Rat)) : α n m /-- Swaps two rows. -/ swapRows {n m : Nat} (mat : α n m) (i j : Nat) : α n m /-- Subtracts `i`-th row multiplied by `coef` from `j`-th row. -/ subtractRow {n m : Nat} (mat : α n m) (i j : Nat) (coef : Rat) : α n m /-- Divides the `i`-th row by `coef`. -/ divideRow {n m : Nat} (mat : α n m) (i : Nat) (coef : Rat) : α n m export UsableInSimplexAlgorithm (setElem getValues ofValues swapRows subtractRow divideRow)
class
Tactic
[ "Mathlib.Init", "Std.Data.HashMap.Basic" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Datatypes.lean
UsableInSimplexAlgorithm
Specification for matrix types over ℚ which can be used in the Gauss Elimination and the Simplex Algorithm. It was introduced to unify dense matrices and sparse matrices.
DenseMatrix (n m : Nat) where /-- The content of the matrix. -/ data : Array (Array Rat)
structure
Tactic
[ "Mathlib.Init", "Std.Data.HashMap.Basic" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Datatypes.lean
DenseMatrix
Structure for matrices over ℚ. So far it is just a 2d-array carrying dimensions (that are supposed to match with the actual dimensions of `data`), but the plan is to add some `Prop`-data and make the structure strict and safe. Note: we avoid using `Matrix` because it is far more efficient to store a matrix as its entries than as function between `Fin`-s.
SparseMatrix (n m : Nat) where /-- The content of the matrix. -/ data : Array <| Std.HashMap Nat Rat
structure
Tactic
[ "Mathlib.Init", "Std.Data.HashMap.Basic" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Datatypes.lean
SparseMatrix
Structure for sparse matrices over ℚ, implemented as an array of hashmaps, containing only nonzero values.
Tableau (matType : Nat → Nat → Type) [UsableInSimplexAlgorithm matType] where /-- Array containing the basic variables' indexes -/ basic : Array Nat /-- Array containing the free variables' indexes -/ free : Array Nat /-- Matrix of coefficients the basic variables expressed through the free ones. -/ mat : matType basic.size free.size
structure
Tactic
[ "Mathlib.Init", "Std.Data.HashMap.Basic" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Datatypes.lean
Tableau
`Tableau` is a structure the Simplex Algorithm operates on. The `i`-th row of `mat` expresses the variable `basic[i]` as a linear combination of variables from `free`.
GaussM (n m : Nat) (matType : Nat → Nat → Type) := StateT (matType n m) Lean.CoreM variable {n m : Nat} {matType : Nat → Nat → Type} [UsableInSimplexAlgorithm matType]
abbrev
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Gauss.lean
GaussM
The monad for the Gaussian Elimination algorithm.
findNonzeroRow (rowStart col : Nat) : GaussM n m matType <| Option Nat := do for i in [rowStart:n] do if (← get)[(i, col)]! != 0 then return i return none
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Gauss.lean
findNonzeroRow
Finds the first row starting from the `rowStart` with nonzero element in the column `col`.
getTableauImp : GaussM n m matType <| Tableau matType := do let mut free : Array Nat := #[] let mut basic : Array Nat := #[] let mut row : Nat := 0 let mut col : Nat := 0 while row < n && col < m do Lean.Core.checkSystem decl_name%.toString match ← findNonzeroRow row col with | none => free := free.push col col := col + 1 continue | some rowToSwap => modify fun mat => swapRows mat row rowToSwap modify fun mat => divideRow mat row mat[(row, col)]! for i in [:n] do if i == row then continue let coef := (← get)[(i, col)]! if coef != 0 then modify fun mat => subtractRow mat row i coef basic := basic.push col row := row + 1 col := col + 1 for i in [col:m] do free := free.push i let ansMatrix : matType basic.size free.size := ← do let vals := getValues (← get) |>.filterMap fun (i, j, v) => if j == basic[i]! then none else some (i, free.findIdx? (· == j) |>.get!, -v) return ofValues vals return ⟨basic, free, ansMatrix⟩
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Gauss.lean
getTableauImp
Implementation of `getTableau` in `GaussM` monad.
getTableau (A : matType n m) : Lean.CoreM (Tableau matType) := do return (← getTableauImp.run A).fst
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/Gauss.lean
getTableau
Given matrix `A`, solves the linear equation `A x = 0` and returns the solution as a tableau where some variables are free and others (basic) variable are expressed as linear combinations of the free ones.
stateLP {n m : Nat} (A : matType n m) (strictIndexes : List Nat) : matType (n + 2) (m + 3) := /- +2 due to shifting by `f` and `z` -/ let objectiveRow : List (Nat × Nat × Rat) := (0, 0, -1) :: strictIndexes.map fun idx => (0, idx + 2, 1) let constraintRow : List (Nat × Nat × Rat) := [(1, 1, 1), (1, m + 2, -1)] ++ (List.range m).map (fun i => (1, i + 2, 1)) let valuesA := getValues A |>.map fun (i, j, v) => (i + 2, j + 2, v) ofValues (objectiveRow ++ constraintRow ++ valuesA)
def
Tactic
[ "Lean.Meta.Basic", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.SimplexAlgorithm", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Gauss" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/PositiveVector.lean
stateLP
Given matrix `A` and list `strictIndexes` of strict inequalities' indexes, we want to state the Linear Programming problem which solution would give us a solution for the initial problem (see `findPositiveVector`). As an objective function (that we are trying to maximize) we use sum of coordinates from `strictIndexes`: it suffices to find the nonnegative vector that makes this function positive. We introduce two auxiliary variables and one constraint: * The variable `y` is interpreted as "homogenized" `1`. We need it because dealing with a homogenized problem is easier, but having some "unit" is necessary. * To bound the problem we add the constraint `x₁ + ... + xₘ + z = y` introducing new variable `z`. The objective function also interpreted as an auxiliary variable with constraint `f = ∑ i ∈ strictIndexes, xᵢ`. The variable `f` has to always be basic while `y` has to be free. Our Gauss method implementation greedy collects basic variables moving from left to right. So we place `f` before `x`-s and `y` after them. We place `z` between `f` and `x` because in this case `z` will be basic and `Gauss.getTableau` produce tableau with nonnegative last column, meaning that we are starting from a feasible point.
extractSolution (tableau : Tableau matType) : Array Rat := Id.run do let mut ans : Array Rat := Array.replicate (tableau.basic.size + tableau.free.size - 3) 0 for h : i in [1:tableau.basic.size] do ans := ans.set! (tableau.basic[i] - 2) <| tableau.mat[(i, tableau.free.size - 1)]! return ans
def
Tactic
[ "Lean.Meta.Basic", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.SimplexAlgorithm", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Gauss" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/PositiveVector.lean
extractSolution
Extracts target vector from the tableau, putting auxiliary variables aside (see `stateLP`).
findPositiveVector {n m : Nat} {matType : Nat → Nat → Type} [UsableInSimplexAlgorithm matType] (A : matType n m) (strictIndexes : List Nat) : Lean.Meta.MetaM <| Array Rat := do /- State the linear programming problem. -/ let B := stateLP A strictIndexes /- Using Gaussian elimination split variable into free and basic forming the tableau that will be operated by the Simplex Algorithm. -/ let initTableau ← Gauss.getTableau B /- Run the Simplex Algorithm and extract the solution. -/ let res ← runSimplexAlgorithm.run initTableau if res.fst.isOk then return extractSolution res.snd else throwError "Simplex Algorithm failed"
def
Tactic
[ "Lean.Meta.Basic", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.SimplexAlgorithm", "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Gauss" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/PositiveVector.lean
findPositiveVector
Finds a nonnegative vector `v`, such that `A v = 0` and some of its coordinates from `strictCoords` are positive, in the case such `v` exists. If not, throws the error. The latter prevents `linarith` from doing useless post-processing.
SimplexAlgorithmException /-- The solution is infeasible. -/ | infeasible : SimplexAlgorithmException
inductive
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
SimplexAlgorithmException
An exception in the `SimplexAlgorithmM` monad.
SimplexAlgorithmM (matType : Nat → Nat → Type) [UsableInSimplexAlgorithm matType] := ExceptT SimplexAlgorithmException <| StateT (Tableau matType) Lean.CoreM variable {matType : Nat → Nat → Type} [UsableInSimplexAlgorithm matType]
abbrev
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
SimplexAlgorithmM
The monad for the Simplex Algorithm.
doPivotOperation (exitIdx enterIdx : Nat) : SimplexAlgorithmM matType Unit := modify fun s : Tableau matType => Id.run do let mut mat := s.mat let intersectCoef := mat[(exitIdx, enterIdx)]! for i in [:s.basic.size] do if i == exitIdx then continue let coef := mat[(i, enterIdx)]! / intersectCoef if coef != 0 then mat := subtractRow mat exitIdx i coef mat := setElem mat i enterIdx coef mat := setElem mat exitIdx enterIdx (-1) mat := divideRow mat exitIdx (-intersectCoef) let newBasic := s.basic.set! exitIdx s.free[enterIdx]! let newFree := s.free.set! enterIdx s.basic[exitIdx]! have hb : newBasic.size = s.basic.size := by apply Array.size_setIfInBounds have hf : newFree.size = s.free.size := by apply Array.size_setIfInBounds return (⟨newBasic, newFree, hb ▸ hf ▸ mat⟩ : Tableau matType)
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
doPivotOperation
Given indexes `exitIdx` and `enterIdx` of exiting and entering variables in the `basic` and `free` arrays, performs pivot operation, i.e. expresses one through the other and makes the free one basic and vice versa.
checkSuccess : SimplexAlgorithmM matType Bool := do let lastIdx := (← get).free.size - 1 return (← get).mat[(0, lastIdx)]! > 0 && (← (← get).basic.size.allM (fun i _ => do return (← get).mat[(i, lastIdx)]! ≥ 0))
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
checkSuccess
Check if the solution is found: the objective function is positive and all basic variables are nonnegative.
chooseEnteringVar : SimplexAlgorithmM matType Nat := do let mut enterIdxOpt : Option Nat := none -- index of entering variable in the `free` array let mut minIdx := 0 for i in [:(← get).free.size - 1] do if (← get).mat[(0, i)]! > 0 && (enterIdxOpt.isNone || (← get).free[i]! < minIdx) then enterIdxOpt := i minIdx := (← get).free[i]! /- If there is no such variable the solution does not exist for sure. -/ match enterIdxOpt with | none => throwThe SimplexAlgorithmException SimplexAlgorithmException.infeasible | some enterIdx => return enterIdx
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
chooseEnteringVar
Chooses an entering variable: among the variables with a positive coefficient in the objective function, the one with the smallest index (in the initial indexing).
chooseExitingVar (enterIdx : Nat) : SimplexAlgorithmM matType Nat := do let mut exitIdxOpt : Option Nat := none -- index of entering variable in the `basic` array let mut minCoef := 0 let mut minIdx := 0 for i in [1:(← get).basic.size] do if (← get).mat[(i, enterIdx)]! >= 0 then continue let lastIdx := (← get).free.size - 1 let coef := -(← get).mat[(i, lastIdx)]! / (← get).mat[(i, enterIdx)]! if exitIdxOpt.isNone || coef < minCoef || (coef == minCoef && (← get).basic[i]! < minIdx) then exitIdxOpt := i minCoef := coef minIdx := (← get).basic[i]! return exitIdxOpt.get! -- such variable always exists because our problem is bounded
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
chooseExitingVar
Chooses an exiting variable: the variable imposing the strictest limit on the increase of the entering variable, breaking ties by choosing the variable with smallest index.
choosePivots : SimplexAlgorithmM matType (Nat × Nat) := do let enterIdx ← chooseEnteringVar let exitIdx ← chooseExitingVar enterIdx return ⟨exitIdx, enterIdx⟩
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
choosePivots
Chooses entering and exiting variables using (Bland's rule)[(https://en.wikipedia.org/wiki/Bland%27s_rule)] that guarantees that the Simplex Algorithm terminates.
runSimplexAlgorithm : SimplexAlgorithmM matType Unit := do while !(← checkSuccess) do Lean.Core.checkSystem decl_name%.toString let ⟨exitIdx, enterIdx⟩ ← choosePivots doPivotOperation exitIdx enterIdx
def
Tactic
[ "Mathlib.Tactic.Linarith.Oracle.SimplexAlgorithm.Datatypes" ]
Mathlib/Tactic/Linarith/Oracle/SimplexAlgorithm/SimplexAlgorithm.lean
runSimplexAlgorithm
Runs the Simplex Algorithm inside the `SimplexAlgorithmM`. It always terminates, finding solution if such exists.
Edge where /-- Source of the edge. -/ src : Nat /-- Destination of the edge. -/ dst : Nat /-- Proof of `atomToIdx[src] ≤ atomToIdx[dst]`. -/ proof : Expr
structure
Tactic
[ "Mathlib.Tactic.Order.CollectFacts" ]
Mathlib/Tactic/Order/Graph/Basic.lean
Edge
An edge in a graph. In the `order` tactic, the `proof` field stores the of `atomToIdx[src] ≤ atomToIdx[dst]`.
Graph := Array (Array Edge)
abbrev
Tactic
[ "Mathlib.Tactic.Order.CollectFacts" ]
Mathlib/Tactic/Order/Graph/Basic.lean
Graph
If `g` is a `Graph`, then for a vertex with index `v`, `g[v]` is an array containing the edges starting from this vertex.
addEdge (g : Graph) (edge : Edge) : Graph := g.modify edge.src fun edges => edges.push edge
def
Tactic
[ "Mathlib.Tactic.Order.CollectFacts" ]
Mathlib/Tactic/Order/Graph/Basic.lean
addEdge
Adds an `edge` to the graph.