f19c4f5cfecdccd65a65f298fd26f17b9d21ad52,tensorforce/core/models/model.py,Model,api_observe,#Model#,1023

Before Change

            deterministic=tf.constant(value=True, dtype=util.tf_dtype(dtype="bool")),
            independent=tf.constant(value=False, dtype=util.tf_dtype(dtype="bool")),
            optimization=tf.constant(value=False, dtype=util.tf_dtype(dtype="bool")),
            timestep=self.timestep[parallel], episode=self.episode[parallel]
        )

        one = tf.constant(value=1, dtype=util.tf_dtype(dtype="long"))

After Change

        )

        with tf.control_dependencies(control_inputs=assertions):
            reward = self.add_summary(
                label=("timestep-reward", "rewards"), name="timestep-reward", tensor=reward
            )
            assignment = self.episode_reward.scatter_nd_add(
                indices=[(parallel,)], updates=[tf.math.reduce_sum(input_tensor=reward, axis=0)]
            )

Instances