Struct LinearPath 

pub struct LinearPath {
    pub contributions: Vec<f32>,
    pub intercept: f32,
    pub logit: f32,
    pub prediction: f32,
    pub probability: Option<f32>,
}

Decision path for linear regression/logistic regression

Fields

contributions: Vec<f32>

Per-feature contributions: coefficient[i] * input[i]

intercept: f32

Bias term contribution

logit: f32

Raw prediction before activation

prediction: f32

Final prediction

probability: Option<f32>

For classification: probability (sigmoid/softmax output)

Implementations

impl LinearPath

pub fn new( contributions: Vec<f32>, intercept: f32, logit: f32, prediction: f32, ) -> Self

Create a new linear path

pub fn with_probability(self, prob: f32) -> Self

Set probability for classification

pub fn top_features(&self, k: usize) -> Vec<(usize, f32)>

Get top k features by absolute contribution

Trait Implementations

impl Clone for LinearPath

fn clone(&self) -> LinearPath

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for LinearPath

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

Formats the value using the given formatter.

impl DecisionPath for LinearPath

fn explain(&self) -> String

Human-readable explanation

fn feature_contributions(&self) -> &[f32]

Feature importance scores (contribution of each feature)

fn confidence(&self) -> f32

Confidence in this decision (0.0 - 1.0)

fn to_bytes(&self) -> Vec<u8>

Compact binary representation

fn from_bytes(bytes: &[u8]) -> Result<Self, PathError>

Reconstruct from binary representation

impl<'de> Deserialize<'de> for LinearPath

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for LinearPath

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.