Documentation

Atlas.TensorCategories.code.GroupoidMultitensor

structure GroupoidMultitensor.FiniteGroupoidData :
Type (max (u + 1) (v + 1))

A finite groupoid presented combinatorially as objects Obj, morphisms Mor, source/target maps, composition (defined when tgt g = src h), identity, and inverse, all subject to the expected groupoid relations.

Instances For

    The groupoid G has a single object (its object set has cardinality 1).

    Instances For

      The number of objects of the finite groupoid G.

      Instances For

        The subtype of morphisms of G between two specified objects.

        Instances For

          The identity-morphism map ObjMor of a groupoid is injective.

          The image of the identity map: the set of identity morphisms, one per object.

          Instances For

            The groupoid has a single connected component contributing to the unit, i.e. only one identity morphism.

            Instances For
              @[reducible, inline]

              The algebra of endomorphisms of the unit object for the groupoid-graded k-vector space category: pointwise scalars indexed by objects of G.

              Instances For

                The dimension of EndUnit equals the number of objects of G.

                The endomorphism algebra of the unit has dimension 1 iff the groupoid has a single object — the criterion separating tensor from multitensor categories.

                If G has a single object, EndUnit k is k-linearly equivalent to k.

                Instances For

                  If G has a single object, EndUnit k is k-algebra-equivalent to k.

                  Instances For

                    If G has more than one object, EndUnit k has dimension greater than one.

                    View a finite group as the single-object groupoid whose morphisms are the group elements.

                    Instances For

                      The pair groupoid on a finite set X: objects are points of X, with a unique morphism between any two of them.

                      Instances For

                        The transformation groupoid X ⋊ G associated with a finite group acting on a finite set: objects are points of X, morphisms are pairs (g, x) from x to g • x.

                        Instances For
                          structure GroupoidMultitensor.GroupoidGradedVecData (k : Type w) [Field k] (G : FiniteGroupoidData) :
                          Type (max v (w + 1))

                          A groupoid-graded vector space: assignment of a k-vector space to each morphism of G. The categories of such gradings model multitensor categories arising from groupoids.

                          Instances For

                            A morphism g is a unit-component iff it equals some identity morphism G.id x.

                            Instances For

                              Grading shift for the dual object: invert the morphism.

                              Instances For
                                @[reducible]

                                Category structure on GroupoidGradedVecData k G: a morphism is a family of k-linear maps between the components indexed by morphisms of G.

                                Instances For
                                  @[implicit_reducible]

                                  Monoidal structure on GroupoidGradedVecData k G whose tensor product convolves gradings along composition in the groupoid.

                                  structure GroupoidMultitensor.MatrixVecData (k : Type w) [Field k] (n : ) :
                                  Type (w + 1)

                                  Matrix-valued vector space data: a k-vector space component i j assigned to each (i, j) index pair in Fin n × Fin n.

                                  Instances For
                                    @[reducible]

                                    Category structure on MatrixVecData k n: morphisms are entrywise linear maps.

                                    Instances For
                                      @[implicit_reducible]

                                      Monoidal structure on MatrixVecData k n using matrix multiplication of components.

                                      structure GroupoidMultitensor.ProductMatrixGroupVecData (k : Type w) [Field k] {m : } (n : Fin m) (G : Fin mType w) [(i : Fin m) → Group (G i)] [(i : Fin m) → Fintype (G i)] [(i : Fin m) → DecidableEq (G i)] :
                                      Type (w + 1)

                                      The product of m matrix-group-graded vector space datas: for each i we have a matrix indexed by Fin (n i) × Fin (n i) further graded by elements of G i.

                                      Instances For
                                        @[reducible]
                                        def GroupoidMultitensor.pmgvCategoryInstance (k : Type w) [Field k] {m : } (n : Fin m) (G : Fin mType w) [(i : Fin m) → Group (G i)] [(i : Fin m) → Fintype (G i)] [(i : Fin m) → DecidableEq (G i)] :

                                        Category structure on ProductMatrixGroupVecData with componentwise linear maps.

                                        Instances For
                                          @[implicit_reducible]
                                          noncomputable instance GroupoidMultitensor.pmgvMonoidalInstance (k : Type w) [Field k] {m : } (n : Fin m) (G : Fin mType w) [(i : Fin m) → Group (G i)] [(i : Fin m) → Fintype (G i)] [(i : Fin m) → DecidableEq (G i)] :

                                          Monoidal structure on ProductMatrixGroupVecData combining matrix multiplication and the group multiplications of the components.

                                          Two monoidal categories C and D are monoidally equivalent if there is an equivalence between them whose forward functor carries a lax monoidal structure.

                                          Instances For

                                            The monoidal category of pair-groupoid-graded vector spaces on X is monoidally equivalent to the matrix vector-space category of size |X|.

                                            theorem GroupoidMultitensor.groupoid_decomposition (k : Type w) [Field k] (G : FiniteGroupoidData) :
                                            ∃ (m : ) (n : Fin m) (autGroup : Fin mType w) (x : (i : Fin m) → Group (autGroup i)) (x_1 : (i : Fin m) → Fintype (autGroup i)) (x_2 : (i : Fin m) → DecidableEq (autGroup i)), AreMonoidallyEquivalent (GroupoidGradedVecData k G) (ggvCategoryInstance k G) (ggvMonoidalInstance k G) (ProductMatrixGroupVecData k n autGroup) (pmgvCategoryInstance k n autGroup) (pmgvMonoidalInstance k n autGroup)

                                            Decomposition theorem: every groupoid-graded vector space category is monoidally equivalent to a finite product of matrix-group-graded vector space categories, one per connected component.

                                            The grading set indexing the (x, y) component subcategory: morphisms in G from y to x.

                                            Instances For
                                              noncomputable def GroupoidMultitensor.fromMathlibGroupoid (X : Type u) [Fintype X] [DecidableEq X] [CategoryTheory.Groupoid X] [(x y : X) → Fintype (x y)] [(x y : X) → DecidableEq (x y)] :

                                              Bridge from a Mathlib Groupoid instance with finite Hom sets to FiniteGroupoidData.

                                              Instances For