Power¶

Power  // Elementwise exponentiation operation


Description¶

Elementwise exponentiation operation.

Inputs¶

Name Element Type Shape
arg0 any any
arg1 same as arg0 same as arg0

Outputs¶

Name Element Type Shape
output same as arg0 same as arg0

Mathematical Definition¶

$\mathtt{output}_{i_0, \ldots, i_{n-1}} = \mathtt{arg0}_{i_0, \ldots, i_{n-1}} ^ {\mathtt{arg1}_{i_0, \ldots, i_{n-1}}}$

Backprop¶

$\begin{split}\overline{\mathtt{arg0}} &\leftarrow \frac{\Delta \cdot \mathtt{arg1}}{\mathtt{arg0}} \\ \overline{\mathtt{arg1}} &\leftarrow \Delta \cdot \mathtt{output} \cdot \log(\mathtt{arg1})\end{split}$

C++ Interface¶

class Power : public ngraph::op::util::BinaryElementwiseArithmetic

Elementwise exponentiation operation.

## Inputs

| | Type | Description | | —— | ——————————— | —————————————————— | | arg0 | $$N[d_1,\dots,d_n]~(n \geq 0)$$ | A tensor of any shape and numeric element type. | | arg1 | $$N[d_1,\dots,d_n]~(n \geq 0)$$ | A tensor of the same shape and element type as arg0. |

## Output

| Type | Description | | ———————- | ————————————————————————————————————– | | $$N[d_1,\dots,d_n]$$ | The tensor $$T$$, where $$T[i_1,\dots,i_n] = \texttt{arg0}[i_1,\dots,i_n]^{\texttt{arg1}[i_1,\dots,i_n]}$$ |

Public Functions

Power(const std::shared_ptr<Node> &arg0, const std::shared_ptr<Node> &arg1)

Constructs an exponentiation operation.

Parameters
• arg0: Node that produces the first input tensor.
• arg1: Node that produces the second input tensor.