Conditional Independence (d-separation) with Networkx in Python

[@deepakkapoor5427]

Can we implement this in pandas dataframe? If yes please provide some reference. Thanks

[@MachineLearningSimulation]

Hey, I assume that you want to check certain data attributes for (conditional) independence? This reminds me of the Chi-Square test. That is not directly what this video was about. Here we are using the architecture of a Directed Graphical Model to determine (conditional) independence in the joint distributions. I assume you are looking for a purely data-driven approach? In the future, I am planning to do videos on the Chi-Square test (also with Pandas). Until then, maybe this could be a first pointer: medium.com/swlh/how-to-run-chi-square-test-in-python-4e9f5d10249d Let me know if I got you correctly :) I would be happy to help

[@syedhasan773]

@@MachineLearningSimulation Hi, I have the same question. I am trying to create a bayesian network using the apyori library. so can I use NetworkX package along with this library? Or lets say can Networkx work with pandas dataframes? Thanks :)