Releases: spike-imperial/FastLAS
v2.2.0
v2.1.0
An extension of FastLAS, including IncrementalLAS, as used in the experiments for the upcoming IJCAI 2022 paper "Search Space Expansion for Efficient Incremental Inductive Logic Programming".
The version adds support for the IncrementalLAS algorithm. To run IncrementalLAS, use the --read-cache and --write-cache flags (if the file for --read-cache does not exist, it will be created by --write-cache).
v2.0.0
The second version of FastLAS, including FastNonOPL, as used in the experiments for the upcoming IJCAI 2021 paper "Scalable Non-observational Predicate Learning in ASP".
This version removes #constant delcarations. Constants must now be declared in the background knowledge. For example #constant(type, c). should be replaced with a background fact type(c).
The version also adds support for the FastNonOPL algorithm. To run FastNonOPL, use the --nopl flag. To run the original FastLAS algorithm, use the --opl flag.
v1.0.4
v1.0.3
The first version of FastLAS, as used in the experiments for the AAAI 2020 paper "FastLAS: Scalable Inductive Logic Programming incorporating Domain-specific Optimisation Criteria".
This release addresses a bug introduced in v1.0.2 where rules were incorrectly identified as being isomorphic. Currently, not all isomorphic rules are now being removed. This is a temporary fix that will be improved in the next release.
v1.0.2
The first version of FastLAS, as used in the experiments for the AAAI 2020 paper "FastLAS: Scalable Inductive Logic Programming incorporating Domain-specific Optimisation Criteria".
This release is more efficient in cases where rules contain many variables.
v1.0.1
The first version of FastLAS, as used in the experiments for the AAAI 2020 paper "FastLAS: Scalable Inductive Logic Programming incorporating Domain-specific Optimisation Criteria".
This release contains minor bug fixes, including the correct reporting of unsatisfiable tasks.
FastLAS version 1.0.0
The first version of FastLAS, as used in the experiments for the AAAI 2020 paper "FastLAS: Scalable Inductive Logic Programming incorporating Domain-specific Optimisation Criteria".