getCovariance()
In statistics, covariance is a measure of the relationship between two random variables. The metric evaluates how much, or to what extent – the variables change together.
If X and Y are two random variables, with means (expected values) $\mu X$ and $\mu Y$, their covariance is as follow:
\[covariance = E[(X - \mu X)(Y - \mu Y)]\]Import
import * as datacook from '@pipcook/datacook';
const { getCovariance } = datacook.Stat;
Syntax
getCovariance(x: Tensor1D | number[], y: Tensor1D | number[]): number
Parameters
| Parameter | type | description |
|---|---|---|
| x | Tensor1D|number[] | first input data of shape (nSamples,) in type of array or tensor |
| y | Tensor1D|number[] | second input data of shape (nSamples,) in type of array or tensor |
Returns
<number> covariance of x and y
Usage
const x = [5, 10, 2, 4, 2];
const y = [2, 8, 7, 6, 1];
const cov = getCovariance(x, y);
console.log(cov);
// 4.9