chisquare_like

chisquare_like(
    x: &Tensor<T>,
    df: T
) -> Result<Tensor<T>, TensorError>

Same as chisquare but the shape will be based on x. Creates a Tensor with values drawn from a chi-square distribution with df degrees of freedom.

Parameters:

x: Input Tensor to derive the shape from

df: Degrees of freedom parameter. Must be positive.

Returns:

Tensor with type T containing random values from the chi-square distribution.

Examples:

use hpt::{error::TensorError, ops::Random, Tensor};

fn main() -> Result<(), TensorError> {
    // Create an initial tensor
    let x = Tensor::<f32>::randn(&[10, 10])?;
    
    // Create a new tensor with same shape as x but with chi-square distribution
    let c = x.chisquare_like(5.0)?;
    println!("{}", c);
    Ok(())
}

Backend Support

Backend	Supported
CPU	✅
Cuda	❌