Pre-release

This release is a pre-release and may not be stable for production use.

PSyKE

PSyKE Logo

Quick links:

Latest Releases

PSyKE 1.0: Compatibility with Python 3.11.x
PSyKE 0.10: New genetic algorithms for knowledge extraction
PSyKE 0.9: Fairness mitigation support for knowedge extractors
PSyKE 0.8: New features: local explainability and counterfactual support
PSyKE 0.7: New SKE algorithms implemented

Intro

PSyKE (Platform for Symbolic Knowledge Extraction) is intended as a library for extracting symbolic knowledge (in the form of logic rule lists) out of sub-symbolic predictors.

More precisely, PSyKE offers a general purpose API for knowledge extraction, and a number of different algorithms implementing it, supporting both classification and regression problems. The extracted knowledge consists of a Prolog theory (i.e., a list of Horn clauses) or an OWL ontology containing SWRL rules.

PSyKE relies on 2ppy (tuProlog in Python) for logic support, which in turn is based on the 2p-Kt logic ecosystem.

Class diagram overview:

PSyKE class diagram

PSyKE is designed around the notion of extractor. More precisely, an Extractor is any object capable of extracting a logic Theory out of a trained sub-symbolic regressor or classifier. Accordingly, an Extractor is composed of (i) a trained predictor (i.e., black-box used as an oracle) and (ii) a set of feature descriptors, and it provides two methods:

extract: returns a logic theory given a dataset;
predict: predicts a value using the extracted rules (instead of the original predictor).

Currently, the supported extraction algorithms are:

CART, straightforward extracts rules from both classification and regression decision trees;
Classification:
- REAL (Rule Extraction As Learning), generates and generalizes rules strarting from dataset samples;
- Trepan, generates rules by inducing a decision tree and possibly exploiting m-of-n expressions;
Regression:
- ITER, builds and iteratively expands hypercubes in the input space. Each cube holds a constant value, that is the estimated output for the samples inside the cube;
- GridEx, extension of the ITER algorithm that produces shorter rule lists retaining higher fidelity w.r.t. the predictor.
- GridREx, extension of GridEx where the output of each hypercube is a linear combination of the input variables and not a constant value.

Users may exploit the PEDRO algorithm, included in PSyKE, to tune the optimal values for GridEx and GridREx hyper-parameters.

We are working on PSyKE to extend its features to encompass explainable clustering tasks, as well as to make more general-purpose the supported extraction algorithms (e.g., by adding classification support to GridEx and GridREx).

Users

End users

PSyKE is deployed as a library on Pypi. It can be installed as Python package by running:

pip install psyke

Requirements

Please refer to the requirements file

Test requirements

skl2onnx
onnxruntime
parameterized

Once installed, it is possible to create an extractor from a predictor (e.g. Neural Network, Support Vector Machine, K-Nearest Neighbours, Random Forest, etc.) and from the data set used to train the predictor.

Note: the predictor must expose a method named predict to be properly used as an oracle.

End users

A brief example is presented in demo.py script in the demo/ folder. Using sklearn's Iris data set we train a K-Nearest Neighbours to predict the correct output class. Before training, we make the dataset discrete. After that we create two different extractors: REAL and Trepan. We output the extracted theory for both extractors.

REAL extracted rules:

iris(PetalLength, PetalWidth, SepalLength, SepalWidth, setosa) :- PetalWidth =< 1.0.
iris(PetalLength1, PetalWidth1, SepalLength1, SepalWidth1, versicolor) :- PetalLength1 > 4.9, SepalWidth1 in [2.9, 3.2].
iris(PetalLength2, PetalWidth2, SepalLength2, SepalWidth2, versicolor) :- PetalWidth2 > 1.6.
iris(PetalLength3, PetalWidth3, SepalLength3, SepalWidth3, virginica) :- SepalWidth3 =< 2.9.
iris(PetalLength4, PetalWidth4, SepalLength4, SepalWidth4, virginica) :- SepalLength4 in [5.4, 6.3].
iris(PetalLength5, PetalWidth5, SepalLength5, SepalWidth5, virginica) :- PetalWidth5 in [1.0, 1.6].

Trepan extracted rules:

iris(PetalLength6, PetalWidth6, SepalLength6, SepalWidth6, virginica) :- PetalLength6 > 3.0, PetalLength6 in [3.0, 4.9].
iris(PetalLength7, PetalWidth7, SepalLength7, SepalWidth7, versicolor) :- PetalLength7 > 3.0.
iris(PetalLength8, PetalWidth8, SepalLength8, SepalWidth8, setosa) :- true.

Developers

Working with PSyKE codebase requires a number of tools to be installed:

Python 3.11
- Python version >= 3.12.x are currently not supported
JDK 11+ (please ensure the JAVA_HOME environment variable is properly configured)
Git 2.20+

Develop PSyKE with PyCharm

To participate in the development of PSyKE, we suggest the PyCharm IDE.

Importing the project

Clone this repository in a folder of your preference using git_clone appropriately
Open PyCharm
Select Open
Navigate your file system and find the folder where you cloned the repository
Click Open

Developing the project

Contributions to this project are welcome. Just some rules:

We use git flow, so if you write new features, please do so in a separate feature/ branch
We recommend forking the project, developing your code, then contributing back via pull request
Commit often
Stay in sync with the develop (or master) branch (pull frequently if the build passes)
Do not introduce low quality or untested code

Issue tracking

If you meet some problems in using or developing PSyKE, you are encouraged to signal it through the project "Issues" section on GitHub.

Metadata

Release files for psyke 1.0.4.dev66

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for psyke 1.0.4.dev66
File	Size	Uploaded
psyke-1.0.4.dev66.tar.gz	74.8 kB	Sep 12, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for psyke 1.0.4.dev66
File	Interpreter	ABI	Platform	Reset
psyke-1.0.4.dev66-py3-none-any.whl 77.4 kB Sep 12, 2026	Python 3	none	any	Details

Total release size: 152.2 kB

Release files / psyke-1.0.4.dev66.tar.gz

Download URL	psyke-1.0.4.dev66.tar.gz
Size	74.8 kB
Tags	Source
SHA-256 checksum How to use checksums	`03a57cb46d86788a0365d8eacfb7b2ab4e2ed7ca68204baf4d229706dfd5248b`
BLAKE2b-256 checksum How to use checksums	`a7d93363696798341355155453b059279b99a900800a0c2203e962730394fcf8`
Upload date	Sep 12, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.11.16`

Release files / psyke-1.0.4.dev66-py3-none-any.whl

Download URL	psyke-1.0.4.dev66-py3-none-any.whl
Size	77.4 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`cf96584aa699f644dd6eec92fc95bf2768ccaf1c797372484d1c93a2772ee72c`
BLAKE2b-256 checksum How to use checksums	`ad7ff6860497df9dedc20413ce695c5cc9f788bf2cd097a2b3ad8d7006a93b60`
Upload date	Sep 12, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.11.16`