7.3.3.1. Custom Solvers

• Setup a Custom Solver

  Setting up a custom solver in MLPro-GT-Native is very simple. There are one method to be defined and other two methods to be optionally defined, as following:

  from mlpro.gt.native.basics import *
  
  class MySolver(GTSolver):
  
      C_NAME      = 'MySolver'
  
      # Optional
      def _init_hyperparam(self, **p_param):
          """
          Implementation specific hyperparameters can be added here. Please follow these steps:
          a) Add each hyperparameter as an object of type HyperParam to the internal hyperparameter
          space object self._hyperparam_space
          b) Create hyperparameter tuple and bind to self._hyperparam_tuple
          c) Set default value for each hyperparameter
  
          Parameters
          ----------
          p_param : Dict
              Further model specific hyperparameters, that are passed through constructor.
          """
  
          ...
  
  
      # Optional
      def _setup_solver(self):
          """
          A method to setup a solver. This needs to be redefined based on each policy, but remains
          optional.
  
          """
  
          ...
  
      # Mandatory
          def _compute_strategy(self, p_payoff:GTPayoffMatrix) -> GTStrategy:
              """
              A method to compute a strategy from the solver. This method needs to be redefined.
  
              Parameters
              ----------
              p_payoff : GTPayoffMatrix
                  Payoff matrix of a specific player.
  
              Returns
              -------
              GTStrategy
                  The computed strategy.
  
              """
  
              ...
  

• Algorithm Checker

  A test script using a unit test to check the developed solvers will be available soon!