Trait DescriptiveStatisticsAlgorithms 

pub trait DescriptiveStatisticsAlgorithms<R: Runtime<DType = DType>>: TensorOps<R> {
    // Required methods
    fn describe(&self, x: &Tensor<R>) -> Result<TensorDescriptiveStats<R>>;
    fn percentile(&self, x: &Tensor<R>, p: f64) -> Result<Tensor<R>>;
    fn iqr(&self, x: &Tensor<R>) -> Result<Tensor<R>>;
    fn skewness(&self, x: &Tensor<R>) -> Result<Tensor<R>>;
    fn kurtosis(&self, x: &Tensor<R>) -> Result<Tensor<R>>;
    fn zscore(&self, x: &Tensor<R>) -> Result<Tensor<R>>;
    fn sem(&self, x: &Tensor<R>) -> Result<Tensor<R>>;

    // Provided method
    fn median(&self, x: &Tensor<R>) -> Result<Tensor<R>> { ... }
}

Descriptive statistics algorithms for tensors.

Provides methods for computing comprehensive statistical summaries of tensor data, including central tendency, dispersion, and shape measures.

Required Methods

fn describe(&self, x: &Tensor<R>) -> Result<TensorDescriptiveStats<R>>

Compute comprehensive descriptive statistics for a 1D tensor.

Returns min, max, mean, variance, std, skewness, and kurtosis as tensors.

fn percentile(&self, x: &Tensor<R>, p: f64) -> Result<Tensor<R>>

Compute the p-th percentile.

Arguments

• x - Input tensor
• p - Percentile to compute (0-100)

fn iqr(&self, x: &Tensor<R>) -> Result<Tensor<R>>

Compute the interquartile range (Q3 - Q1).

fn skewness(&self, x: &Tensor<R>) -> Result<Tensor<R>>

Compute skewness (Fisher’s definition).

fn kurtosis(&self, x: &Tensor<R>) -> Result<Tensor<R>>

Compute excess kurtosis (Fisher’s definition).

fn zscore(&self, x: &Tensor<R>) -> Result<Tensor<R>>

Compute z-scores (standardized values).

fn sem(&self, x: &Tensor<R>) -> Result<Tensor<R>>

Compute standard error of the mean.

Provided Methods

fn median(&self, x: &Tensor<R>) -> Result<Tensor<R>>

Compute the median (50th percentile).

Dyn Compatibility

This trait is dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementations on Foreign Types

impl DescriptiveStatisticsAlgorithms<CpuRuntime> for CpuClient

fn describe( &self, x: &Tensor<CpuRuntime>, ) -> Result<TensorDescriptiveStats<CpuRuntime>>

fn percentile( &self, x: &Tensor<CpuRuntime>, p: f64, ) -> Result<Tensor<CpuRuntime>>

fn iqr(&self, x: &Tensor<CpuRuntime>) -> Result<Tensor<CpuRuntime>>

fn skewness(&self, x: &Tensor<CpuRuntime>) -> Result<Tensor<CpuRuntime>>

fn kurtosis(&self, x: &Tensor<CpuRuntime>) -> Result<Tensor<CpuRuntime>>

fn zscore(&self, x: &Tensor<CpuRuntime>) -> Result<Tensor<CpuRuntime>>

fn sem(&self, x: &Tensor<CpuRuntime>) -> Result<Tensor<CpuRuntime>>

Implementors