Enum Dist 

pub enum Dist {
    Gaussian,
    Uniform,
    Sparse90,
    LowRank,
    Outlier,
    Quantized,
    Coherent,
}

Synthetic input distributions. Each is chosen to make a different sketch pop, so the miner can be validated against ground truth.

Variants

Gaussian

N(0,1) — the null hypothesis (dense, bell-shaped, no structure).

Uniform

U(-1,1) — dense, flat histogram.

Sparse90

90% exact zeros — unstructured sparsity → sparse-GEMM candidate.

LowRank

Rank-4 product — low effective rank → factored-matmul candidate.

Outlier

Gaussian with ~1% ×30 spikes → per-channel outliers → SmoothQuant/AWQ.

Quantized

Gaussian snapped to 0.5 steps → spiky histogram → quantize/LUT.

Coherent

Row random-walk — adjacent rows cohere → within-call sequence structure → delta-compute candidate (proves the adj_sumsq detector).

Implementations

impl Dist

pub const ALL: [Dist; 7]

pub fn name(self) -> &'static str

Trait Implementations

impl Clone for Dist

fn clone(&self) -> Dist

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for Dist

impl Debug for Dist

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Eq for Dist

impl PartialEq for Dist

fn eq(&self, other: &Dist) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for Dist