AFAICT this is a library that implements the 'do calculus' as described in Pearl's 'Book of why'. What I recall from that book: - RCTs aren't the only way to

AFAICT this is a library that implements the 'do calculus' as described in Pearl's 'Book of why'. What I recall from that book: - RCTs aren't the only way to find causality - you DO need a model of your situation, that shows all the possible causal chains - if you have this model then, using data, you can calculate the influence of each part using the 'do calculus' - the ideas have been met with resistance