Symbolic regression example with Python visualization
In this tutorial we find a formula for the first prime numbers and visualize the result with Python.
Machine learning without black boxes.
TuringBot discovers mathematical formulas that predict variables. Avanced modelling made easy.
Features
The software is based on a technique called Symbolic Regression. Here is what it can do:
Easily create sophisticated models
Creating mathematical models is a difficult task, but TuringBot makes it effortless: all you have to do is load your data in TXT or CSV format, select which column should be predicted and which ones should be used as input, and then simply start the search.
Both regression and classification models can be generated by choosing from the several search metrics available.
Simplicity first
A good model is one that is both accurate and simple. That is why TuringBot will give you a list of the best formulas of all possible sizes, discarding complicated ones that do not perform better than simpler alternatives.
The optional cross validation feature can be activated to help select the formula with the best trade-off between complexity and accuracy, allowing overfit solutions to be ruled out in a straightforward way.
Machine learning without black boxes
Symbolic models are simple and easy to understand. They show exacly which variables are being used and how. This is a clear advantage over more commonly used methods like neural networks, which are complete black boxes.
The models that TuringBot finds do not only spill out the correct numbers. They offer insight and understanding into your data.
Try TuringBot today!
No credit card required.
Blog
Want to see TuringBot in action?
Check out our official blog.
Using R to visualize a Symbolic Regression model
Here we show how a symbolic regression model can be visualized using R and the ggplot2 library.
Using Symbolic Regression to predict rare events
Learn how to classify credit card fraud in a real world dataset using Symbolic Regression.