Update

Author: ikeuchi-screen

CONTRIBUTING.md

@@ -25,7 +25,7 @@ In addition, we add the following guidelines:
 Coding style is checked with flake8.
 
 ``` sh
-flake8 lingam --count --ignore=E203,E741,C901 --max-line-length=127 --statistics
+flake8 lingam --count --ignore=E203,E501,E741,C901 --max-line-length=127 --statistics
 ```
 
 ## Documentation
@@ -35,9 +35,7 @@ When adding a new feature to lingam, you also need to document it in the referen
 To build the documentation, you use Sphinx. Run the following commands to install Sphinx and its extensions.
 
 ``` sh
-pip install sphinx
-pip install sphinxcontrib-napoleon
-pip install sphinx_rtd_theme
+pip install -r ./docs/requirements-doc.txt
 ```
 
 Then you can build the documentation in HTML format locally:

README.md

@@ -10,24 +10,16 @@ LiNGAM is a new method for estimating structural equation models or linear Bayes
 ## Requirements
 
 * Python3
-
 * numpy
-
 * scipy
-
 * scikit-learn
-
 * graphviz
-
 * statsmodels
-
-- networkx
-
-- pandas
-
-- itertools
-
-- semopy
+* networkx
+* pandas
+* itertools
+* semopy
+* autograd
 
 
 ## Installation
@@ -150,6 +142,10 @@ Should you use this package for performing **GroupLiNGAM algorithm**, we kindly
 
 * Y. Kawahara, K. Bollen, S. Shimizu and T. Washio. **GroupLiNGAM: Linear non-Gaussian acyclic models for sets of variables.** Arxiv preprint arXiv:1006.5041, 2010. [[PDF]](https://arxiv.org/abs/1006.5041)
 
+Should you use this package for performing **ABIC-LiNGAM algorithm**, we kindly request you to cite the following paper:
+
+* Y. Morinishi and S. Shimizu. **Differentiable causal discovery of linear non-Gaussian acyclic models under unmeasured confounding.** Transactions on Machine Learning Research (TMLR), 2025. [[PDF]](https://openreview.net/forum?id=HR7MFlW73I)
+
 
 ### Causality and prediction