MCPcopy Create free account
hub / github.com/InternRobotics/NavDP / act

Method act

baselines/ddppo/pointnav_network.py:76–109  ·  view source on GitHub ↗

Infers action to take towards the given (rho, theta) based on depth vision. Args: observations (Union["TensorDict", Dict]): A dictionary containing (at least) the following: - "depth" (torch.float32): Depth image tensor (N, H, W, 1).

(
        self,
        observations: Union["TensorDict", Dict],
        masks: Tensor,
        deterministic: bool = False,
    )

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 6

forwardMethod · 0.45
forwardMethod · 0.45
ddppo_step_pointgoalFunction · 0.45
actMethod · 0.45
forwardMethod · 0.45
forwardMethod · 0.45

Calls 1

move_obs_to_deviceFunction · 0.85

Tested by

no test coverage detected