docs

Author: felixleopoldo

docs/render_alg_docs.py

@@ -162,23 +162,42 @@ def info_to_small_table():
     # This is the module part
 
     module_str = "\n\n"
+
+    meta_description = ""
+    if info["meta_description"] == info["title"]:
+        meta_description = content.replace("\n", " ").replace("\r", " ").replace("\t", " ").replace('"', "'")
+    else:
+        meta_description = info["meta_description"]
+
     module_str += """
     .. meta::
         :title: {} 
         :description: {}
-    """.format(info["title"], content.replace("\n", " ").replace("\r", " ").replace("\t", " ").replace('"', "'")) 
-    #         :keywords: causal discovery, causal discovery algorithm, Benchpress, graphical models, probabilistic graphical models, structure learning, benchmarking, graph estimation, graph learning, graph structure, structure learning algorithms, {}, {}
+    """.format(info["title_full"], meta_description) 
+    
     module_str += "\n\n"
     module_str += ".. _"+p.name+": \n\n"
+    #module_str +="" + p.name +" - " + info["title_full"] +" \n"
+    #module_str +="-"*len(p.name) + "-"*4 + "-"*len(info["title_full"]) + "-"*4 + "\n"
+    
+
     module_str +="" + p.name +" \n"
-    module_str +="-"*len(p.name) + "-"*4 + "\n"
+    module_str +="*"*len(p.name) + "*"*4 + "\n"
+
+
 
     module_str += "\n"
-    module_str += ".. rubric:: "+ info["title"]    
+    #module_str += ".. rubric:: "+ info["title_full"]    
+    #module_str += info["title_full"] + "\n"    
+    #module_str +="-"*len(info["title_full"]) + "-"*4 + "\n"
     module_str += "\n\n"
     module_str += info_to_table(info, p)
     module_str += "\n\n"
-    module_str += ".. rubric:: Description"    
+    module_str +="" + info["title_full"] +" \n"
+    module_str +="-"*len(info["title_full"]) + "-"*4 + "\n"
+    
+    
+    #module_str += ".. rubric:: " + info["title_full"]    
     if content != "":    
         module_str += "\n\n"
         module_str += content

docs/source/conf.py

@@ -85,8 +85,8 @@
 # a list of builtin themes.
 #
 html_theme = 'sphinx_rtd_theme'
-html_title = 'Benchpress causal discovery platform'
-html_short_title = 'Benchpress causal discovery platform'
+html_title = 'Benchpress'
+html_short_title = 'Benchpress'
 
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,

docs/source/index.rst

@@ -5,8 +5,8 @@
 
 
 .. meta::
-    :title: Benchpress: Scalable Open-Source Software for Causal Discovery
-    :description: Benchpress: A flexible, platform-independent open-source tool for developing, executing, and benchmarking causal discovery algorithms. Scales effortlessly across cores, servers, clusters, grids, and cloud environments for seamless performance.
+    :title: Benchpress: Scalable Open-Source Causal Discovery
+    :description: Benchpress is a platform-independent open-source software to develop, execute, and benchmark causal discovery algorithms.
 .. toctree::
     :hidden:
     :glob:
@@ -29,7 +29,7 @@
 
 .. toctree::
     :hidden:
-    :maxdepth: 3
+    :maxdepth: 1
     :name: Modules
     :caption: Modules
 
@@ -66,17 +66,17 @@
 
 ------------------------
 
-##################################
-Benchpress
-##################################
+#############################################################
+Scalable and open-source causal discovery
+#############################################################
 
 Describing the relationship between the variables in a study domain and modelling
 the data generating mechanism is a fundamental problem in many empirical sciences.
 `Probabilistic graphical models <https://en.wikipedia.org/wiki/Graphical_model>`_ are one common approach to tackle the problem. 
-Learning the graphical structure for such models (sometimes called causal discovery) is computationally challenging and a fervent
+Learning the graphical structure for such models, referred to as **structure learning** or **causal discovery**, is computationally challenging and a fervent
 area of current research with a plethora of algorithms being developed. 
 To facilitate the access to the different methods we present Benchpress, a scalable and platform-independent `Snakemake <https://snakemake.github.io/>`_ workflow to **run**, **develop**, and to create reproducible **benchmarks**
-of **structure learning algorithms** for probabilistic graphical models. 
+of **structure learning algorithms** for probabilistic graphical models.
 Benchpress is interfaced via a simple `JSON <https://www.json.org/json-en.html>`_-file, which makes it accessible for all users, while the code is
 designed in a fully modular fashion to enable researchers to contribute additional methodologies. 
 Benchpress  provides an interface to a large number of state-of-the-art

docs/source/structure_learning_algorithms/athomas_jtsamplers.rst

@@ -9,9 +9,9 @@
 .. _athomas_jtsamplers: 
 
 athomas_jtsamplers 
-----------------------
+**********************
+
 
-.. rubric:: Chordal graph samplers
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ athomas_jtsamplers
 
 
 
-.. rubric:: Description
+Chordal graph samplers 
+--------------------------
+
 
 Abstract :footcite:p:`10.2307/43304539`: Full Bayesian computational inference for model determination in undirected graphical models is currently restricted to decomposable graphs or other special cases, except for small-scale problems, say up to 15 variables. In this paper we develop new, more efficient methodology for such inference, by making two contributions to the computational geometry of decomposable graphs. The first of these provides sufficient conditions under which it is possible to completely connect two disconnected complete subsets of vertices, or perform the reverse procedure, yet maintain decomposability of the graph. The second is a new Markov chain Monte Carlo sampler for arbitrary positive distributions on decomposable graphs, taking a junction tree representing the graph as its state variable. 
 

docs/source/structure_learning_algorithms/bdgraph.rst

@@ -9,9 +9,9 @@
 .. _bdgraph: 
 
 bdgraph 
------------
+***********
+
 
-.. rubric:: BDgraph
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ bdgraph
 
 
 
-.. rubric:: Description
+BDgraph 
+-----------
+
 
 
 

docs/source/structure_learning_algorithms/bidag_itsearch.rst

@@ -9,9 +9,9 @@
 .. _bidag_itsearch: 
 
 bidag_itsearch 
-------------------
+******************
+
 
-.. rubric:: Iterative MCMC
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ bidag_itsearch
 
 
 
-.. rubric:: Description
+Iterative MCMC 
+------------------
+
 
 This is a hybrid score-based optimisation technique based on Markov chain Monte Carlo
 schemes :footcite:t:`doi:10.1080/10618600.2021.2020127`. The algorithm starts from a skeleton obtained

docs/source/structure_learning_algorithms/bidag_order_mcmc.rst

@@ -9,9 +9,9 @@
 .. _bidag_order_mcmc: 
 
 bidag_order_mcmc 
---------------------
+********************
+
 
-.. rubric:: Order MCMC
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ bidag_order_mcmc
 
 
 
-.. rubric:: Description
+Order MCMC 
+--------------
+
 
 This technique relies on a Bayesian perspective on structure learning, where the score of a DAG
 is defined as its posterior distribution. To overcome the limitation of simple structure-based

docs/source/structure_learning_algorithms/bidag_partition_mcmc.rst

@@ -9,9 +9,9 @@
 .. _bidag_partition_mcmc: 
 
 bidag_partition_mcmc 
-------------------------
+************************
+
 
-.. rubric:: Partition MCMC
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ bidag_partition_mcmc
 
 
 
-.. rubric:: Description
+Partition MCMC 
+------------------
+
 
 Abstract: 
 Acyclic digraphs are the underlying representation of Bayesian networks, a widely used class of probabilistic graphical models. Learning the underlying graph from data is a way of gaining insights about the structural properties of a domain. Structure learning forms one of the inference challenges of statistical graphical models. Markov chain Monte Carlo (MCMC) methods, notably structure MCMC, to sample graphs from the posterior distribution given the data are probably the only viable option for Bayesian model averaging. Score modularity and restrictions on the number of parents of each node allow the graphs to be grouped into larger collections, which can be scored as a whole to improve the chain’s convergence. Current examples of algorithms taking advantage of grouping are the biased order MCMC, which acts on the alternative space of permuted triangular matrices, and nonergodic edge reversal moves. Here, we propose a novel algorithm, which employs the underlying combinatorial structure of DAGs to define a new grouping. As a result convergence is improved compared to structure MCMC, while still retaining the property of producing an unbiased sample. Finally, the method can be combined with edge reversal moves to improve the sampler further. Supplementary materials for this article are available online.

docs/source/structure_learning_algorithms/bnlearn_fastiamb.rst

@@ -9,9 +9,9 @@
 .. _bnlearn_fastiamb: 
 
 bnlearn_fastiamb 
---------------------
+********************
+
 
-.. rubric:: Fast IAMB
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ bnlearn_fastiamb
 
 
 
-.. rubric:: Description
+Fast IAMB 
+-------------
+
 
 From bnlearn: a variant of IAMB which uses speculative stepwise forward selection to reduce the number of conditional independence tests.
 

docs/source/structure_learning_algorithms/bnlearn_gs.rst

@@ -9,9 +9,9 @@
 .. _bnlearn_gs: 
 
 bnlearn_gs 
---------------
+**************
+
 
-.. rubric:: Grow-shrink
 
 .. list-table:: 
 
@@ -35,7 +35,9 @@ bnlearn_gs
 
 
 
-.. rubric:: Description
+Grow-shrink 
+---------------
+
 
 The grow-shrink (GS) algorithm is based on the Markov blanket of the nodes in a DAG. For
 a specific node, the Markov blanket it the set of nodes which conditioning upon renders it