Python TFPolicyGraph.extra_compute_action_fetches Beispiele

Beispiel #1

Datei: dqn_policy_graph.py Projekt: wesleykerr/ray

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self), **{
         "q_values": self.q_values,
     })

Beispiel #2

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self),
                 **self._extra_action_fetches)

Beispiel #3

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self),
                 **{"vf_preds": self.vf})

Beispiel #4

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self),
                 **{"behaviour_logits": self.model.outputs})

Beispiel #5

Datei: appo_policy_graph.py Projekt: zbarry/ray

 def extra_compute_action_fetches(self):
     out = {"behaviour_logits": self.model.outputs}
     if not self.config["vtrace"]:
         out["vf_preds"] = self.value_function
     return dict(TFPolicyGraph.extra_compute_action_fetches(self), **out)

Beispiel #6

 def extra_compute_action_fetches(self):
     return dict(
         TFPolicyGraph.extra_compute_action_fetches(self), **{
             SampleBatch.VF_PREDS: self.value_function,
             BEHAVIOUR_LOGITS: self.logits
         })

Beispiel #7

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self),
                 **{SampleBatch.VF_PREDS: self.vf})

Beispiel #8

Datei: ppo_policy_graph.py Projekt: robertnishihara/ray

 def extra_compute_action_fetches(self):
     return dict(
         TFPolicyGraph.extra_compute_action_fetches(self), **{
             "vf_preds": self.value_function,
             "logits": self.logits
         })

Beispiel #9

Datei: appo_policy_graph.py Projekt: robertnishihara/ray

 def extra_compute_action_fetches(self):
     out = {"behaviour_logits": self.model.outputs}
     if not self.config["vtrace"]:
         out["vf_preds"] = self.value_function
     return dict(TFPolicyGraph.extra_compute_action_fetches(self), **out)

Beispiel #10

Datei: human_trainer.py Projekt: vladfi1/DeepSmash

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self),
                 **{"behavior_logp": self.sampled_logp})

Beispiel #11

Datei: dqn_policy_graph.py Projekt: robertnishihara/ray

 def extra_compute_action_fetches(self):
     return dict(
         TFPolicyGraph.extra_compute_action_fetches(self), **{
             "q_values": self.q_values,
         })

Beispiel #12

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self), **{
         "vf_preds": self.value_function,
         "logits": self.logits
     })

Beispiel #13

Datei: a3c_tf_policy_graph.py Projekt: robertnishihara/ray

 def extra_compute_action_fetches(self):
     return dict(
         TFPolicyGraph.extra_compute_action_fetches(self),
         **{"vf_preds": self.vf})

Beispiel #14

Datei: vtrace_policy_graph.py Projekt: zhouhh2017/ray

 def extra_compute_action_fetches(self):
     return dict(TFPolicyGraph.extra_compute_action_fetches(self),
                 **{BEHAVIOUR_LOGITS: self.model.outputs})

Beispiel #15

Datei: vtrace_policy_graph.py Projekt: robertnishihara/ray

 def extra_compute_action_fetches(self):
     return dict(
         TFPolicyGraph.extra_compute_action_fetches(self),
         **{"behaviour_logits": self.model.outputs})