Args: box_coord (tuple): a tuple containing x0, y0, x1, y1 coordinates, where x0 and y0 are the coordinates of th

Args: segment: numpy array of shape Nx2, containing all the points in the polygon. color: color of the polygon. Refer

Args: segment: numpy array of shape Nx2, containing all the points in the polygon. color: color of the polygon. Refer

embed goal+instr language

Click the specified DOM element.

Click at coordinates (left, top).

Send keystrokes to the focused element.

Generates masks for the given image. Arguments: image (np.ndarray): The image to generate masks for, in HWC uint8 format.

Get a copy of the default config.

Function for getting closest teach category by querying LLM

Gets a base embedding for one dimension with sin and cos intertwined

Gets a base embedding for one dimension with sin and cos intertwined

Return initial positions of agents in an episode :param episode: instance of class Episode in dataset.py

object_name: Object category of object to find object_phrase: grounding phrase to help locate the object parent_object: paren

Gets object info from simulator directly

reprojects centroid of ID to 2D based on current position from navigation

Zigzag scan of a matrix Argument is a two-dimensional matrix of any size, not strictly a square one. Function returns

load action as a list of tokens from task_json

Given the arguemnts, load and initialize the configs. Args: args (argument): arguments includes `shard_id`, `num_shards`,

load numericalized language from task_json

Apply non-maximum suppression at test time to avoid detecting too many overlapping bounding boxes for a given object. Args: boxes: (te

Lower text and remove punctuation, articles and extra whitespace.

Return Euclidean distance between a given object and the nearest agent in the sim.

Args: boxes (Boxes, RotatedBoxes or ndarray): either a :class:`Boxes`, or an Nx4 numpy array of XYXY_ABS format f

Args: boxes (Boxes, RotatedBoxes or ndarray): either a :class:`Boxes`, or an Nx4 numpy array of XYXY_ABS format f

Parse the following arguments for a default parser for PySlowFast users. Args: shard_id (int): shard id for the current machine. Star

Assumes object is in view and picks it up. Returns: bool: True if action successful, otherwise False.

Assumes object is in view and picks it up. Returns: bool: True if action successful, otherwise False.

Puts down the object in hand on a nearby receptacle. Note: Assumes object is in hand and agent wants to free its hand.

Reset environment

Reset the environment. :param options: options for the environment. The current supported options are: - "storage_state":

Args: img: same as in __init__

Looping through games and write out status at the end

Human in the loop learning. Runs script and refines it if an execution failure occurs based on feedback.

Search for target object near another receptacle_target object max_rec_to_search: if there are many of this receptacle in memory, how

multi_scale_features is from the key encoder for skip-connection memory_readout is from working/long-term memory sensory is t

Read in an episode from file. game_fn: the game logfile to read task_idx: the task in the game logfile to read episod

setup logger for training and testing. Args: save_dir(str): location to save log file distributed_rank(int): device rank when mult

Get info of all product in a given order page. Example output: [ { "name": "Kellogg's Special K Protein Bars, Meal R

Helper function to synchronize (barrier) among all processes when using distributed training

Return a single unified embedding from the four embeddings (task, obs, act, visual).