Feyn Documentation

Feyn Documentation

  • Learn
  • Guides
  • Tutorials
  • API Reference
  • FAQ

Learn how to use Feyn and the QLattice

Feyn is a Python module built on top of the QLattice for supervised machine learning using symbolic regression.

Whether you're an absolute beginner or a seasoned veteran, this is the place to be for links and resources to learn more about how to use Feyn in practice.


Just starting out?

Here's a few good places to start exploring

Using Feyn

Regression and classification

Using Feyn

Auto run

Get the details about Auto Run

Understanding Auto Run

Categorical values

Learn how Feyn handles categoricals

Understanding categoricals

Tutorial: Titanic Classification

Learn how to use Feyn with the Titanic Dataset

Titanic Classification

Tutorial: Concrete Strength Regression

Learn how to use Feyn with the Concrete Strength Dataset

oncrete Strength Regression

Explainability

Learn about features that go into explainability

Response Plots

Learn about interpreting your models with the response plot

Automatically Plot Partials

Tutorial: Partial Plot

A simple regression case using the 1D response plot with quantiles to understand the model

Constrained 1D Response Plot

Tutorial: Liver Cancer in Plasma

Picking out multicollinearity with the QLattice

Exploring multicollinearity with the QLattice

Hypothesis generation

Here's some cases that go into working with multiple hypotheses

Querying for Models

How to use the query language to sample specific models

Using the Query Language

Tutorial: Rewriting Models

Substituting inputs in a model with correlated alternatives

Rewriting models

Tutorial: Covid mRNA Vaccine Degradation

A life science case that demonstrates expanding and constraining model complexity to understand relationships better

Covid mRNA Vaccine Degradation

Custom Workflows

Did you know that training in Feyn is composed of primitive operations? Here's some examples of how to use them to customise your workflow

Using the primitives

Build your own workflow with the primitive operations

Using the primitives

Tutorial: Plotting loss during training

Use the primitives to customise the plots during training

Loss graph during training

Tutorial: Wine Quality Regression

Predict the alcohol levels of wine using the primitive operations

Wine Quality Regression

Using Feyn in production

Here's some ideas on what to do after you've gotten a model you're happy with

Loading and Saving models

How to save and load fitted models

Loading and saving models

Creating Sympy Expressions

Converting a model to a mathematical expression

Converting models to sympy expressions

Subscribe to get news about Feyn and the QLattice.

You can opt out at any time, and you can read our privacy policy here.

Copyright © 2024 Abzu.ai - Feyn license: CC BY-NC-ND 4.0
Feyn®, QGraph®, and the QLattice® are registered trademarks of Abzu®