updates navigation mapping inputs rgb: rgb of current frame depth: depth map current frame action_successful: whether

Return a valid action object with type GOTO_URL.

empty out the object / clear the object from other objects

search_object: object category string or list of category strings - stop navigation and return object info when found

Render the trajectory

Stop the environment session.

Return the playwright action

Executes manipulation action subgoal_name Args: subgoal_name (string): action name to execute. landmark_name (Intera

Removes small disconnected regions and holes in a mask. Returns the mask and an indicator of if the mask has been modified.

Execute an Interaction - a formatted action - and add it to the current episode :param interaction: instance of class Interaction def

Args: binary_mask (ndarray): numpy array of shape (H, W), where H is the image height and W is the image width. E

Args: binary_mask (ndarray): numpy array of shape (H, W), where H is the image height and W is the image width. E

compute encodings for all tokens using a normal flat encoder

Gets navigable points around the agent at various yaw angles (bins navigable points around the agent based on yaw angle around object center)

Create configs and perform basic setups.

key: C*N values: dict of values (C*N), object ids are used as keys shrinkage: 1*N selection: C*N supposed_bu

Return a segmentation mask highlighting object(s) in an egocentric image :param agent_id: the agent whose image needs to be highlight

Navigates to the object.

Predict the next action given the observation

Picks up the interaction object and places it on the specified landmark. Args: landmark_name (InteractionObject): Landmark t

Compute of box in coordinate format (min, max).

Position camera specified by camera_id such that object obj is visible; Used to set target object view for TEACh data collection inte

Given degrees y in [0, 359], return the closest cardinal direction as a tuple (x_dir, z_dir) in {-1, 0, 1}^2 :param y: Input angle in

Translate AI2-THOR errorMessage field into something that can be shown as prompts to annotators for TEACh data collection

Depending on the brightness_factor, gives a lighter or darker color i.e. a color with less or more saturation than the original color

Depending on the brightness_factor, gives a lighter or darker color i.e. a color with less or more saturation than the original color