Loading vagen/env/frozenlake/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -19,8 +19,8 @@ class FrozenLakeEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: id_fields=["is_slippery", "size", "p", "render_mode", "max_actions_per_step", "min_actions_to_succeed","format_reward"] Loading vagen/env/navigation/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -22,8 +22,8 @@ class NavigationEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling max_objects_in_state: int = 10 use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: """Generate a unique identifier for this configuration.""" Loading vagen/env/primitive_skill/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -18,8 +18,8 @@ class PrimitiveSkillEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: Loading vagen/env/sokoban/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -18,8 +18,8 @@ class SokobanEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: id_fields = ["dim_room", "max_steps", "num_boxes", "render_mode", "min_actions_to_succeed", "max_actions_per_step"] Loading Loading
vagen/env/frozenlake/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -19,8 +19,8 @@ class FrozenLakeEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: id_fields=["is_slippery", "size", "p", "render_mode", "max_actions_per_step", "min_actions_to_succeed","format_reward"] Loading
vagen/env/navigation/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -22,8 +22,8 @@ class NavigationEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling max_objects_in_state: int = 10 use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: """Generate a unique identifier for this configuration.""" Loading
vagen/env/primitive_skill/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -18,8 +18,8 @@ class PrimitiveSkillEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: Loading
vagen/env/sokoban/env_config.py +2 −2 Original line number Diff line number Diff line Loading @@ -18,8 +18,8 @@ class SokobanEnvConfig(BaseEnvConfig): # configs for process reward for grounding and world modeling use_state_reward: bool = False grounding_reward_weight: float = 0.5 worldmodeling_reward_weight: float = 0.5 grounding_reward_weight: float = 2.0 worldmodeling_reward_weight: float = 2.0 def config_id(self) -> str: id_fields = ["dim_room", "max_steps", "num_boxes", "render_mode", "min_actions_to_succeed", "max_actions_per_step"] Loading
