ed288758872ae7246c8ed078d0cb78f910ad4ed5,tensorforce/agents/dqfd_agent.py,DQFDAgent,observe,#DQFDAgent#Any#Any#,217

Before Change

        if self.timestep >= self.first_update and self.timestep % self.update_frequency == 0:
            for _ in xrange(self.repeat_update):
                batch = self.demo_memory.get_batch(batch_size=self.demo_batch_size, next_states=True)
                self.model.demonstration_update(
                    states={name: np.stack((batch["states"][name], batch["next_states"][name])) for name in batch["states"]},
                    internals=batch["internals"],
                    actions=batch["actions"],
                    terminal=batch["terminal"],
                    reward=batch["reward"]
                )

    def import_demonstrations(self, demonstrations):
        
        Imports demonstrations, i.e. expert observations. Note that for large numbers of observations,

After Change

        // TODO Where are these parameters?
        if self.timestep >= self.first_update and self.timestep % self.update_frequency == 0:
            for _ in xrange(self.repeat_update):
                self.model.demonstration_update()

    def import_demonstrations(self, demonstrations):
        
        Imports demonstrations, i.e. expert observations. Note that for large numbers of observations,

Instances