Did you know ... Search Documentation:
Packs (add-ons) for SWI-Prolog

Package "phil"

Title:Learning Hierarchical Probabilistic Logic Programs parameters with gradient descent and Expectation Maximization
Rating:Not rated. Create the first rating!
Latest version:1.0.0
SHA1 sum:fa65e4a7bf652c028b347267459fdf8141fb9ac5
Author:Nguembang Fadja Arnaud <arnaud.nguembafadja@unife.it>
Download URL:https://github.com/ArnaudFadja/phil/releases/*.zip
Requires:auc
matrix

Reviews

No reviews. Create the first review!.

Details by download location

VersionSHA1#DownloadsURL
1.0.00c27b74145440089e2cfc65c7b2268e2c2abc1b21https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
285f18741d64c43ecdf67ceef46d48b2e8f6f92d1https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
3099725365d3c8b5a94e2176f5db001b8ae783311https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
378cd9c4894add8f504236535c8b454750e27b2f1https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
3793f566b8f549c637ebf92781cd12e7087eb70f1https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
3c3827a8145f35002b95c7e7c2d45aeda739b5121https://github.com/ArnaudFadja/phil.git
465d12ffd83a3e55a1cc301352b9dbec7baceaf71https://github.com/ArnaudFadja/phil.git
560b4d8162a0a5168d6bfc9f11e63276ddfb3f341https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
582c491bdc0ff9fdc4c7e84ef66b29763ec1b8fb2https://github.com/ArnaudFadja/phil.git
58c568958318184b4ea604e88b7872a2c36b7f1c1https://github.com/ArnaudFadja/phil.git
62d597de353f5b9ae53bff10c16a13a681b077e84https://github.com/ArnaudFadja/phil.git
714be43ab2e6bbc37f5da1366d7201e90421bfce3https://github.com/ArnaudFadja/phil.git
716d18d6e784eb7c90722d6bb089ad352976dd1f1https://github.com/ArnaudFadja/phil.git
7c89100d18f85e021be2dc8a982249dd4bcfed1e2https://github.com/ArnaudFadja/phil.git
7e985bc122a691def124a27491885fa91b2b7cf71https://github.com/ArnaudFadja/phil.git
8aba88d3422b82d3153259bba27fd469dc5fde071https://github.com/ArnaudFadja/phil.git
91d94c15d84b17143e8b98674190cf5ac35cfb491https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
92a602b4177f5d36bebecae99c1debc13634126b1https://github.com/ArnaudFadja/phil.git
a6694b9bf28c1ad41dec2e9d677e0d59e75b6a721https://github.com/ArnaudFadja/phil.git
b5ed8cd2b6f9ee804c0918a2b32dcda4ce8d606e2https://github.com/ArnaudFadja/phil.git
bc8f8c7f26309d9ce8d20907cb3cbdefa9c473063https://github.com/ArnaudFadja/phil.git
bec507ec3b591f63b996c99af4bde7764b99f4c81https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
c0faa8504f604af3a3aad5b62cef14b44932b5c82https://github.com/ArnaudFadja/phil.git
ccdd38febd256435b77d6a53f6d2598fd4afc6392https://github.com/ArnaudFadja/phil.git
d556bb2bf8c7560026b4311162fea2cf4f3c841d4https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
dab41ba51b50e7a2aa057e15ea30c97cae6dedb31https://github.com/ArnaudFadja/phil.git
db419c6d0b4f2eaabd59efc5da27806a387d3c571https://github.com/ArnaudFadja/phil/archive/v1.0.0.zip
e25b627c35fc9179b2db98f15df521d3f4626b132https://github.com/ArnaudFadja/phil.git
f7d0cecedb5827f7de404da543379af204e47dc43https://github.com/ArnaudFadja/phil.git
fa65e4a7bf652c028b347267459fdf8141fb9ac53https://github.com/ArnaudFadja/phil.git

phil

phil is a parameter learning algorithm that learns the parameters of Hierarchical Probabilistic Logic Programs (HPLP) applying gradient descent (dphil) or Expectation Maximization (emphil).

Installation

This is a SWI-Prolog (http://www.swi-prolog.org/) pack.

It can be installed with pack_install/1

$ swipl
?- pack_install(phil).

The pack uses a foreign library and contains the library binaries for 32 and 64 bits Linux and 32 and 64 bits Windows. If you want to recompile the foreign library you can use

?- pack_rebuild(phil).

On 32 and 64 bits Linux this should work out of the box. On 32 and 64 bits Windows the library must be rebuilt by hand. First run pack_rebuild(phil). This typically fails but produces the file buildenv.sh in the root folder. You can modify this file looking at the example files buildenvmingw32.sh and buildenvmingw64.sh. Then you can run

$ source buildenv.sh
$ source configure
$ make install

Requirements

It requires packs auc and matrix:

$ swipl
?- pack_install(auc).
?- pack_install(matrix).

Example of use

Datasets are available in pack cplint_datasets (https://github.com/ArnaudFadja/phil_datasets) Install the phil_datasets with pack_install/1

$ swipl
?- pack_install(phil_datasets).

Then

$ cd <pack>/phil/prolog/
$ swipl
?- [uwcsedeep].
?- induce_par([ai,graphics,language,systems,theory],P),test(P,[ai],LL,AUCROC,ROC,AUCPR,PR).

Contents of pack "phil"

Pack contains 489 files holding a total of 64.1M bytes.