Expand description
linfa-linear aims to provide pure Rust implementations of popular linear regression algorithms.
§The Big Picture
linfa-linear is a crate in the linfa ecosystem, an effort to create a toolkit for classical Machine Learning
implemented in pure Rust, akin to Python’s scikit-learn.
§Current state
linfa-linear currently provides an implementation of the following regression algorithms:
- Ordinary Least Squares
- Generalized Linear Models (GLM)
- Isotonic
§Examples
There is an usage example in the examples/ directory. To run, use:
$ cargo run --features openblas --example diabetes
$ cargo run --example glmStructs§
- Fitted
Isotonic Regression - A fitted isotonic regression model which can be used for making predictions.
- Fitted
Linear Regression - A fitted linear regression model which can be used for making predictions.
- Isotonic
Regression - An isotonic regression model.
- Linear
Regression - An ordinary least squares univariate linear regression model.
- Tweedie
Regressor - Generalized Linear Model (GLM) with a Tweedie distribution
- Tweedie
Regressor Params - The set of hyperparameters that can be specified for the execution of the Tweedie Regressor.
- Tweedie
Regressor Valid Params - The set of hyperparameters that can be specified for the execution of the Tweedie Regressor.
Enums§
- Linear
Error - An error when modeling a Linear algorithm
- Link
- Link functions used by GLM