Skip to Content
DocumentationFunctionalBinary Opspow_op

pow_op

View the code on GitHub 

Structs

Struct: Pow

Fields

Methods

fwd(arg0: Array, arg1: Array) -> Array
Raises the first array to the power of the second array element-wise.
Args

  • arg0: Array The first input array.

  • arg1: Array The second input array.

Returns
  • Array - The element-wise result of raising arg0 to the power of arg1.

Examples:

a = Array([[1, 2], [3, 4]])
b = Array([[5, 6], [7, 8]])
result = pow_to(a, b)
print(result)

This function supports

  • Broadcasting.
  • Automatic differentiation (forward and reverse modes).
  • Complex valued arguments.
jvp(primals: List[Array], tangents: List[Array]) -> Array
Compute Jacobian-vector product for array exponentiation.
Args

  • primals: List[Array] Primal input arrays.

  • tangents: List[Array] Tangent vectors.

Returns
  • Array - Array: Jacobian-vector product.

Note: Implements forward-mode automatic differentiation for exponentiation. The result represents how the output changes with respect to infinitesimal changes in the inputs along the directions specified by the tangents.

See Also:

pow_vjp: Reverse-mode autodiff for exponentiation.

vjp(primals: List[Array], grad: Array, out: Array) -> List[Array]
Compute vector-Jacobian product for array exponentiation.
Args

  • primals: List[Array] Primal input arrays.

  • grad: Array Gradient of the output with respect to some scalar function.

  • out: Array The output of the forward pass.

Returns
  • List[Array] - List[Array]: Gradients with respect to each input.

Note: Implements reverse-mode automatic differentiation for exponentiation. Returns arrays with shape zero for inputs that do not require gradients.

See Also:

pow_jvp: Forward-mode autodiff for exponentiation.

binary_simd_op(arg0_real: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)], arg1_real: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)], arg0_imag: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)], arg1_imag: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)]) -> Tuple[SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)], SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)]]
Low-level function to raise a complex number to a complex power represented as SIMD vectors.
Args

  • arg0_real: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)] The real part of the complex number.

  • arg1_real: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)] The real part of the power.

  • arg0_imag: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)] The imaginary part of the complex number.

  • arg1_imag: SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)] The imaginary part of the power.

Returns
  • Tuple[SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)], SIMD[float32, nelts[::DType]().__mul__(2).__floordiv__(2)]] - The real and imaginary parts of the complex number raised to the complex power as a tuple.
__call__(mut curr: Array, args: List[Array])
Raises the first array to the power of the second array element-wise and stores the result in the current array (curr). The function assumes that the shape and data of the args are already set up. If the shape and data of the current array (curr) is not set up, the function will compute the shape based on the shapes of the args and set up the data accordingly.
Args

  • curr: Array The current array, must be mutable.

  • args: List[Array] The two arrays to raise to the power.

Functions

pow_to

pow_to(arg0: Array, arg1: Array) -> Array
Raises the first array to the power of the second array element-wise.
Args

  • arg0: Array The first input array.

  • arg1: Array The second input array.

Returns
  • Array - The element-wise result of raising arg0 to the power of arg1.

Examples:

a = Array([[1, 2], [3, 4]])
b = Array([[5, 6], [7, 8]])
result = pow(a, b)
print(result)

This function supports

  • Broadcasting.
  • Automatic differentiation (forward and reverse modes).
  • Complex valued arguments.
Last updated on
mul_opsub_op