Struct SaeRowJetChannels 

pub struct SaeRowJetChannels {
    pub n_rows: usize,
    pub k: usize,
    pub p: usize,
    pub first: Vec<f64>,
    pub second: Vec<f64>,
}

One row’s order-≤2 reconstruction jet channels, flattened row-major: first[a*p + c] = ∂ẑ_c/∂p_a and second[(a*K + b)*p + c] = ∂²ẑ_c/∂p_a∂p_b, where K = n_primaries and p = out_dim. This is exactly what the CPU fill_reconstruction_channels_from_program writes into the per-row first[a][c] / second[a][b][c] tensors (with sqrt_row_w = 1; the caller applies the per-row loss weight, which is linear and commutes with the GN contraction).

Fields

n_rows: usize

k: usize

p: usize

first: Vec<f64>

n_rows * K * p

second: Vec<f64>

n_rows * K * K * p

Trait Implementations

impl Clone for SaeRowJetChannels

fn clone(&self) -> SaeRowJetChannels

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SaeRowJetChannels

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

Formats the value using the given formatter.

impl PartialEq for SaeRowJetChannels

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for SaeRowJetChannels