XHAIL — Symbolic Learning Playground

Load example

program.lp15 ln

% Background knowledge
bird(a). bird(b). bird(c).
penguin(d).
bird(X) :- penguin(X).

% Mode declarations
#modeh flies(+bird).
#modeb penguin(+bird).
#modeb not penguin(+bird).

% Examples
#example flies(a).
#example flies(b).
#example flies(c).
#example not flies(d).

Deduction depth10

Syntax reference▼

#modeh flies(+bird).Head mode — predicate to learn

#modeb penguin(+bird).Body mode — available conditions

#modeh flies(+bird) :1-3 =2 @1.Mode with min-max, weight, priority

#example flies(a).Positive example

#example not flies(d).Negative example

#example flies(a) =5 @2.Weighted/prioritised example

#display flies/1.Filter output to this predicate

fact(a). r(X) :- body(X).Background knowledge (ASP)

+type = input variable · #type = ground constant · -type = output variable

Pipeline output

2mspenguins

10benchmarks

v0.1on PyPI

$pip install xhail

Load an example or write your own program, then press ⌘ ↵ to run.

◈ Symbolic & interpretable⚡ ASP-powered⬡ ILP / Ray (2009)