Intro

grad

sciann.utils.grad()

Computes gradient tensor of functional object f.

Arguments

  • f: Functional object.
  • ys: layer name for ys to differentiate.
  • xs: layer name for xs to be differentiated w.r.t.
  • order: order of differentiation w.r.t. xs - defaulted to 1.

Returns

A new functional object.

diag_grad

sciann.utils.diag_grad()

Computes diag of gradient tensor of functional object f.

Arguments

  • f: Functional object.
  • ys: layer name for ys to differentiate.
  • xs: layer name for xs to be differentiated w.r.t.
  • order: order of differentiation w.r.t. xs - defaulted to 1.

Returns

A new functional object.

div

sciann.utils.div(other)

Element-wise division applied to the Functional objects.

Arguments

  • f: Functional object.
  • other: A python number or a tensor or a functional object.

Returns

A Functional.

radial_basis

sciann.utils.radial_basis(ci, radii)

Apply radial_basis function to x element-wise.

Arguments

  • xs: List of functional objects.
  • ci: Center of basis functional (same length as xs).
  • radii: standard deviation or radius from the center.

Returns

A new functional object.

sin

sciann.utils.sin()

Computes sin of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

asin

sciann.utils.asin()

Computes asin of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

cos

sciann.utils.cos()

Computes cos of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

acos

sciann.utils.acos()

Computes acos of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

tan

sciann.utils.tan()

Computes tan of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

atan

sciann.utils.atan()

Computes atan of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

tanh

sciann.utils.tanh()

Computes tanh of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

exp

sciann.utils.exp()

Computes exp of x element-wise.

Arguments

  • x: Functional object.

Returns

A new functional object.

pow

sciann.utils.pow(b)

Element-wise exponentiation applied to the Functional object.

Arguments

a, b: pow(a,b) Note that at least one of them should be of type Functional.

Returns

A Functional.

add

sciann.utils.add(other)

Element-wise addition applied to the Functional objects.

Arguments

  • f: Functional object.
  • other: A python number or a tensor or a functional object.

Returns

A Functional.

sub

sciann.utils.sub(other)

Element-wise subtraction applied to the Functional objects.

Arguments

  • f: Functional object.
  • other: A python number or a tensor or a functional object.

Returns

A Functional.

mul

sciann.utils.mul(other)

Element-wise multiplication applied to the Functional objects.

Arguments

  • f: Functional object.
  • other: A python number or a tensor or a functional object.

Returns

A Functional.

div

sciann.utils.div(other)

Element-wise division applied to the Functional objects.

Arguments

  • f: Functional object.
  • other: A python number or a tensor or a functional object.

Returns

A Functional.