Graphicness #
Here we study graphic and cographic matroids.
Column of a node-edge incidence matrix is either all 0,
or has exactly one +1 entry, exactly one -1 entry, and all other elements 0.
Equations
Instances For
Matrix is called graphic iff it is a node-edge incidence matrix of a (directed) graph.
Equations
- A.IsGraphic = ∀ (y : n), IsIncidenceMatrixColumn fun (x : m) => A x y
Instances For
theorem
Matrix.IsGraphic.elem_in_signTypeCastRange
{m : Type u_1}
{n : Type u_2}
[DecidableEq m]
{A : Matrix m n ℚ}
(hA : A.IsGraphic)
(i : m)
(j : n)
:
Every element of a graphic matrix is 1, 0, or -1.
theorem
Matrix.IsGraphic.col_zero_or_two_nonzeros
{m : Type u_1}
{n : Type u_2}
[DecidableEq m]
{A : Matrix m n ℚ}
(hA : A.IsGraphic)
(y : n)
:
Column of a node-edge incidence matrix has either zero or two non-zero entries.
theorem
Matrix.IsGraphic.isTotallyUnimodular
{α : Type u_1}
[DecidableEq α]
{X Y : Set α}
{A : Matrix ↑X ↑Y ℚ}
(hA : A.IsGraphic)
:
Node-edge incidence matrix is totally unimodular.
theorem
Matroid.IsGraphic.isRegular
{α : Type u_1}
[DecidableEq α]
{M : Matroid α}
(hM : M.IsGraphic)
:
Graphic matroid is regular.