spike-imperial/fastlas

FastLAS

FastLAS is a system for Inductive Logic Programming, originally presented at AAAI 2020. This repository contains instructions for downloading and installing FastLAS and reproducing the results in the paper.

FastLAS2 is the latest version of the FastLAS system. It allows running both the original FastLAS algorithm and the new FastNonOPL algorithm, which is to be presented at IJCAI 2021. To run the original FastLAS (now known as FastLAS1), use the flag --opl; alternatively, use the flag --nopl to run FastNonOPL.

Any questions regarding FastLAS should be sent to [email protected].

fastlas's Issues

Unable to Run "policy_learning" Task with Precompiled FastLAS2 and Slow Execution with FastLAS v1.0.4

Hello!

I am experiencing issues with running the policy_learning task. Initially, I tried using the precompiled FastLAS2, but it failed to execute this task. As an alternative, I switched to the precompiled FastLAS v1.0.4. While most tasks run successfully with this version, I am encountering a significant long-time execution. Notably, I've been running ./full_cross_validation.rb amazon accept cov for 24 hours, but I still haven't obtained the result from even the first round. I'm curious to know if this is considered normal.

Environment:

• CPU: Intel(R) Xeon(R) Gold 6242 CPU @ 2.80GHz
• CPU(s): 64, Thread(s) per core: 2, Core(s) per socket: 16
• RAM: 125GB
• Ubuntu 20.04
• Ruby 2.7
• Clingo 5.3.0

compilation of FastLAS2

Hi!

I would like to compile FastLAS2 on my machine, but get the following error:

$ make                                                                                                                                                                                                                                                                            [dcefbc9]
[  4%] Building CXX object CMakeFiles/FastLAS_LIB.dir/.parser.cpp.o
cc1plus: error: too many filenames given; type ‘cc1plus --help’ for usage
cc1plus: fatal error: CMakeFiles/FastLAS_LIB.dir/.parser.cpp.d: No such file or directory
compilation terminated.
make[2]: *** [CMakeFiles/FastLAS_LIB.dir/build.make:87: CMakeFiles/FastLAS_LIB.dir/.parser.cpp.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:111: CMakeFiles/FastLAS_LIB.dir/all] Error 2
make: *** [Makefile:91: all] Error 2

Do you know how I can fix that?

Thanks!

Unknown token: '#'

Whenever I try to run FastLAS on the CAVIAR examples (or any other example) it returns either UNSATISFIABLE% or Unknown token: '#'.

Could someone help me?
Example:

FastLAS --opl FastLAS1/data/CAVIAR/files/las__fold_0_FastLAS

how to compute the fn of chunking task?