a41f2ff4c1437f0b61e76265c31bdec71be0556f,softlearning/algorithms/sac.py,SAC,get_diagnostics,#SAC#Any#Any#Any#Any#,358

Before Change

                 self.global_step),
                feed_dict))

        diagnostics = OrderedDict({
            "Q-avg": np.mean(Q_values),
            "Q-std": np.std(Q_values),
            "Q_loss": np.mean(Q_losses),
            "policy_loss": np.mean(policy_losses),
            "alpha": alpha,
        })

        policy_diagnostics = self._policy.get_diagnostics(
            batch["observations"])
        diagnostics.update({

After Change

        

        feed_dict = self._get_feed_dict(iteration, batch)
        diagnostics = self._session.run(self._diagnostics_ops, feed_dict)

        diagnostics.update(OrderedDict([
            (f"policy/{key}", value)
            for key, value in

Instances