You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Dealing with terms should be more general. In the current design, terms in C++ receive a vector. This should change to a list, and more over, using C++ classes:
This way terms can be generated with a set of arbitrary arguments. Moreover, this could be ported to a C++ header-only library for computing sufficient statistics on networks (represented as list of source/target).
On the R side, functions with the same name of the term should be defined such that those can be called within R and set up all the requirements.
Formulas need to be more regeneral. Here is a proposed set of steps:
Extract the RHS of the formula
Match terms by using something like "(edge|ctriad|triad|...)(\(.*\))?"
If only the first part matched, then add () and make a call, otherwise make the call.
The text was updated successfully, but these errors were encountered:
Dealing with terms should be more general. In the current design, terms in C++ receive a vector. This should change to a list, and more over, using C++ classes:
This way terms can be generated with a set of arbitrary arguments. Moreover, this could be ported to a C++ header-only library for computing sufficient statistics on networks (represented as list of source/target).
On the R side, functions with the same name of the term should be defined such that those can be called within R and set up all the requirements.
Formulas need to be more regeneral. Here is a proposed set of steps:
"(edge|ctriad|triad|...)(\(.*\))?"
The text was updated successfully, but these errors were encountered: