Turn your data into formulas that actually make sense

Predicting numerical values from input variables can be challenging. Simple models like linear or polynomial regression often fail to capture complex patterns, while complex machine learning methods can be difficult to interpret and trust.

TuringBot uses symbolic regression to discover clear mathematical formulas that describe your data. The result: models that are accurate, robust, and explainable.

Given input variables like x and z, it can automatically find a relationship such as y = f(x,z).

How does it work?

TuringBot implements a technique called Symbolic Regression. It tries to combine a set of base functions into simple formulas that accurately predict a target variable as a function of one or more input variables. The base functions offered by the program are the following:

• Arithmetic: addition, multiplication, division
• Trigonometric: sin, cos, tan, asin, acos, atan
• Exponential: exp, log, log2, sqrt, pow
• Hyperbolic: sinh, cosh, tanh, asinh, acosh, atanh
• Logical: smaller, greater, equal, different, logical_or, logical_and
• History: delay, moving_average
• Other: abs, floor, ceil, round, sign, mod, gamma, erf

What is optimized is the formula itself, and not just the numerical constants of some assumed model.

The program uses TXT or CSV files as input, which may contain an arbitrary number of columns. It can be executed both interactively through its powerful graphical interface or in an automated way from the command line.

Here is an example of an input file that you can use: input.txt.

Why choose TuringBot?

• Proven track record: TuringBot has been successfully employed in academic publications across a wide range of fields (see below).
• High performance: Powered by a nontrivial algorithm and written in optimized C++, it delivers significantly higher performance than alternatives built on Python or JIT-compiled languages like Julia.
• Easy setup: It is a simple executable that you can install in a few minutes. No Python dependencies, no Docker, no Conda, no virtual environments. This makes it easy to get started and focus on your work.
• For developers and integrators: Need to integrate the program into your workflow? It includes a full command-line interface and a Python library for seamless integration into CI/CD pipelines, Airflow, or any automated workflow.
• Active development: TuringBot's development began in 2019, and version 1.0 was launched to the public in February 2020. Since then, the program has been continually updated in response to user feedback, introducing new features, optimizations, bug fixes, and quality-of-life improvements.

What can TuringBot be used for?

If your problem involves predicting a number as a function of other numbers, then you can apply TuringBot to it. Just save the data in TXT or CSV format, load it in the program, and start the search.

To give a few concrete examples:

• Predict the price of a house as a function of its characteristics (area, number of bedrooms, age, etc): A regression model example and how to generate it.
• Detect fraudulent credit card transactions based on anonymized features: Using Symbolic Regression to predict rare events.

Note that the second example is a classification problem. This is not an issue: just find formulas that output 0 or 1 depending on the category.

A decision boundary found with symbolic regression. Tutorial

What makes TuringBot so general is that many different search metrics are included, allowing models with different goals to be generated. Those include:

• RMS error
• Classification accuracy
• Correlation coefficient
• Maximum error
• Mean error
• Mean relative error
• Maximum relative error
• F-score
• Nash-Sutcliffe efficiency
• Binary cross-entropy
• Matthews correlation

Is TuringBot free?

TuringBot is free forever for datasets up to 50 rows and 3 input variables—no credit card or signup required. Just download and start discovering formulas. For larger datasets and unlimited variables, paid licenses are available on the Pricing page.

Academic publications

Used in 35+ academic publications across Physics, Astrophysics, Engineering, Biology, and Finance.

Some publications that use TuringBot are:

This list is constantly growing and is probably incomplete. If your paper is not shown, please email it to us and we will add it to the list.

Community

• TuringBot Forum: a growing community where you can get help, ask questions, and connect with others who are equally interested in the software.

Documentation

Check out the detailed documentation for the software, where you can find:

• Command-line usage examples
• Python usage examples

The documentation provides straightforward, easy-to-follow examples that will help you get started quickly.

FAQ: How does TuringBot compare to Eureqa?

Both TuringBot and Eureqa are implementations of Symbolic Regression, but the algorithms used by each are completely different. Eureqa is based on genetic programming, while TuringBot is based on Simulated Annealing.

Eureqa was acquired by DataRobot, an enterprise machine learning platform, and is no longer commercially available.

A 2020 paper has shown that TuringBot performs noticeably better than Eureqa on a variety of Physics-inspired problems (arXiv:2010.11328). In this paper, TuringBot even managed to solve problems for which Eureqa could not find a solution at all.