d35509ef4fed2e588adb9285ee1c9bd0eb16582a,softlearning/algorithms/sql.py,SQL,_update_target,#SQL#Any#,320

Before Change

        tau = tau or self._tau

        source_params = self._Q.get_weights()
        target_params = self._Q_target.get_weights()
        self._Q_target.set_weights([
            tau * source + (1.0 - tau) * target
            for source, target in zip(source_params, target_params)
        ])

After Change

    def _update_target(self, tau=None):
        tau = tau or self._tau

        for Q, Q_target in zip(self._Qs, self._Q_targets):
            source_params = Q.get_weights()
            target_params = Q_target.get_weights()
            Q_target.set_weights([
                tau * source + (1.0 - tau) * target
                for source, target in zip(source_params, target_params)
            ])

    def _do_training(self, iteration, batch):
        Run the operations for updating training and target ops.

        feed_dict = self._get_feed_dict(batch)

Instances