TorchLean API

Docs HomeInstallationGuideExamplesGraphsTools

FloatLib.Numerics.Reduction

Collection reductions #

Scalar formats differ in their accumulation rules, but collection-shaped operations face the same structural failures. Keeping those failures here lets binary dot products, posit quires, and future reduction APIs report them without depending on one another.

source
inductive FloatLib.Numerics.ReductionError :
Type

Why a collection reduction could not be evaluated.

  • lengthMismatch (leftLength rightLength : Nat) : ReductionError

    The two input collections contain different numbers of elements.

Instances For
    source
    def FloatLib.Numerics.instReprReductionError.repr :
    ReductionError → Nat → Std.Format
    Instances For
      source
      @[instance_reducible]
      instance FloatLib.Numerics.instReprReductionError :
      Repr ReductionError
      source
      def FloatLib.Numerics.instDecidableEqReductionError.decEq (x✝ x✝¹ : ReductionError) :
      Decidable (x✝ = x✝¹)
      Instances For
        source
        @[instance_reducible]
        instance FloatLib.Numerics.instDecidableEqReductionError :
        DecidableEq ReductionError