Learning

.github/workflows/publish.yml

@@ -0,0 +1,16 @@
+name: Deploy mdBook
+
+on:
+  push:
+    branches: [ "main" ]
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v2
+      - uses: XAMPPRocky/[email protected]
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}

.gitignore

@@ -5,3 +5,5 @@ pyschlandals/target
 .vscode/
 pyschlandals/pipytoken
 .vimspector.json
+*.cnf
+doc/book

Cargo.toml

@@ -12,17 +12,24 @@ license = "AGPL-3.0"
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
 
 [dependencies]
-rustc-hash = "1.1.0"
-clap = { version = "3.0.0-beta.4", features = ["derive"] }
-assert_float_eq = "1.1.3"
-rand = "0.8.5"
-nalgebra = "0.31.4"
-rug = { version = "1.18.0", features = ["float", "integer"] }
-sysinfo = "0.28.1"
-search_trail = "0.1.2"
-peak_alloc = "0.2.0"
+rustc-hash = ">=1.1.0"
+clap = { version = ">=4.4.7", features = ["derive"] }
+assert_float_eq = ">=1.1.3"
+rand = ">=0.8.5"
+nalgebra = ">=0.31.4"
+rug = { version = ">=1.18.0", features = ["float", "integer"] }
+sysinfo = ">=0.28.1"
+search_trail = ">=0.1.2"
+peak_alloc = ">=0.2.0"
+chrono = ">=0.4.31"
+rayon = ">=1.8.0"
+bitvec = ">=1.0.1"
+tch = { version = ">=0.14.0", optional = true }
 
 [dev-dependencies]
 criterion = "0.4.0"
 paste = "1.0.12"
 tempfile = "3"
+
+[features]
+tensor = ["dep:tch"]

doc/book.toml

@@ -0,0 +1,9 @@
+[book]
+authors = ["Alexandre Dubray", "Lucile Dierckx"]
+language = "en"
+multilingual = false
+src = "src"
+title = "The Schlandals Book"
+
+[output.html]
+mathjax-support = true

doc/src/README.md

@@ -0,0 +1,30 @@
+# Introduction
+
+Schlandals is a state-of-the-art _Projected Weighted Model Counter_ specialized for probabilistic inference over discrete probability distributions.
+Currently, there are known modelization for the following problems
+  - Computing the marginal probabilities of a variable in a Bayesian Network
+  - Computing the probability that two nodes are connected in a probabilistic graph
+  - Computing the probability of [ProbLog](https://github.com/ML-KULeuven/problog) programs
+
+For more information on how to use Schlandals and its mechanics, check [the documentation](https://aia-uclouvain.github.io/schlandals/doc).
+You can cite Schlandals using the following bibtex entry
+```
+@InProceedings{dubray_et_al:LIPIcs.CP.2023.15,
+  author =	{Dubray, Alexandre and Schaus, Pierre and Nijssen, Siegfried},
+  title =	{{Probabilistic Inference by Projected Weighted Model Counting on Horn Clauses}},
+  booktitle =	{29th International Conference on Principles and Practice of Constraint Programming (CP 2023)},
+  pages =	{15:1--15:17},
+  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
+  ISBN =	{978-3-95977-300-3},
+  ISSN =	{1868-8969},
+  year =	{2023},
+  volume =	{280},
+  editor =	{Yap, Roland H. C.},
+  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
+  address =	{Dagstuhl, Germany},
+  URL =		{https://drops.dagstuhl.de/opus/volltexte/2023/19052},
+  URN =		{urn:nbn:de:0030-drops-190520},
+  doi =		{10.4230/LIPIcs.CP.2023.15},
+  annote =	{Keywords: Model Counting, Bayesian Networks, Probabilistic Networks}
+}
+```

doc/src/SUMMARY.md

@@ -0,0 +1,16 @@
+# Summary
+
+[Introduction](README.md)
+[Installation](install.md)
+
+- [Schlandals Modelization](modelization/README.md)
+    - [The Language](modelization/language.md)
+    - [Encoding Bayesian Networks](modelization/bn.md)
+    - [Encoding Probabilistic Graphs](modelization/pg.md)
+    - [Encoding ProbLog Programs](modelization/problog.md)
+- [Probabilistic Inference](inference/README.md)
+    - [Search solver](inference/search.md)
+    - [Compiler](inference/compilation.md)
+- [Learning Distributions Parameters](learning/README.md)
+    - [Learning Settings](learning/settings.md)
+- [Python Interface](python.md)

doc/src/inference/README.md

@@ -0,0 +1,5 @@
+# Probabilistic Inference
+
+Schlandals provides exact and \\( \epsilon \\)-bounded approximate inference methods.
+It can be used either as a DPLL-style search-based solver or as a compiler that outputs an arithmetic circuit.
+Beware that, currently; the compiler is just a search-solver saving its trace. Hence, you should use the search solver if you do not need to keep the circuits for further computation.