Welcome to the concision wiki!

Concision aims to be a complete machine-learning framework for Rust.

One focus of the project was to provide a ParamsBase object that allows for the definition of a weight / bias pair. The implementation serves as a natural extension of the ndarray crate by defining automatically defining the bias tensor to be the smaller dimension:

use ndarray::{ArrayBase, Dimension, RawData};

pub struct ParamsBase<S, D, A = <S as RawData>::Elem>
where
    D: Dimension,
    S: RawData<Elem = A>,
{
    pub bias: ArrayBase<S, D::Smaller, A>,
    pub weights: ArrayBase<S, D, A>,
}