8426825800af370988e50f5105ffce53f384de91,tf_agents/policies/random_tf_policy_test.py,RandomTFPolicyTest,testInfoSpec,#RandomTFPolicyTest#Any#,188

Before Change

                        tf.constant(np.log(1. / 3), shape=[batch_size]))

  def testInfoSpec(self, dtype):
    PolicyInfo = collections.namedtuple(  // pylint: disable=invalid-name
        "PolicyInfo",
        ("log_probability", "predicted_rewards"))
    // Set default empty tuple for all fields.
    PolicyInfo.__new__.__defaults__ = ((),) * len(PolicyInfo._fields)

    action_spec = [

After Change

    tf.nest.assert_same_structure(action_spec, batched_action_step.action)
    self.assertEqual((batch_size, 2, 3,), batched_action_step.action[0].shape)
    self.assertEqual((batch_size, 1, 2,), batched_action_step.action[1].shape)
    tf.nest.assert_same_structure(info_spec, batched_action_step.info)
    self.assertEqual((batch_size, 1,), batched_action_step.info[0].shape)
    self.assertEqual((batch_size, 1,), batched_action_step.info[1].shape)

Instances