Library for graphical models of decision making, based on pgmpy and networkx
This release fixes the failing linting tests by update types to make optional types explicit.
This release significantly improves the efficiency of the MACID to EFG transformation for the get_spe() function.
This release adds methods for computing mixed Nash equilibria and mixed subgame perfect equilibria (for an arbitrary number of players) using the pygambit library.
Backwards compatibility Notes:
Note that this breaks backwards compatibility by removing the get_all_pure_ne
, get_all_pure_ne_in_sg
, and get_all_pure_spe
class methods from the MACID class. This is because pure variants of NE and SPE can be found by simply selecting the "enumpure" solver as an argument in the new get_ne
, get_ne_in_sg
, and get_spe
MACID class methods.
This version:
add the method CausalBayesianNetwork.sample()
This release breaks backwards compatibility by forcing the arguments for CPDs to match the case of the names of the parent variables. Previously, the lowercase version of the parent names were used, which lead to less intuitive model specification and more complex and brittle code.
Later versions of pgmpy create bugs in PyCID
In this release, we provide: