MCPcopy Create free account
hub / github.com/YeWR/EfficientZero / representation

Method representation

config/atari/model.py:541–547  ·  view source on GitHub ↗
(self, observation)

Source from the content-addressed store, hash-verified

539 return policy, value
540
541 def representation(self, observation):
542 encoded_state = self.representation_network(observation)
543 if not self.state_norm:
544 return encoded_state
545 else:
546 encoded_state_normalized = renormalize(encoded_state)
547 return encoded_state_normalized
548
549 def dynamics(self, encoded_state, reward_hidden, action):
550 # Stack encoded_state with a game specific one hot encoded action

Callers

nothing calls this directly

Calls 1

renormalizeFunction · 0.90

Tested by

no test coverage detected