Loads a module dictionary from local file or url. Args: filename (str): name of saved module dictionary

Render one frame. Optionally specify dt in seconds, specify None to use wallclock. Specify physics_dt and physics_substeps to decouple the ph

Saves the current module dictionary. Args: filename (str): name of output file

Puts the model to the device. Args: device (device): pytorch device

Convenience function to set settings. Args: setting (str): string representing the setting being changed value: new v

Returns the current USD stage.

Starts the editor physics simulation

Evaluates the occupancy values for the points. Args: p (tensor): points c (tensor): encoded feature volumes

Normalize coordinate to [0, 1] for unit cube experiments Args: p (tensor): point padding (float): conventional padding paramter

Computes the Intersection over Union (IoU) value for two sets of occupancy values. Args: occ1 (tensor): first set of occupancy value

Normalize coordinate to [0, 1] for unit cube experiments. Corresponds to our 3D model Args: x (tensor): coordinate reso

Specify the physics step size to use when simulating, default is 1/60. Note that a physics scene has to be in the stage for this to do anythin

Encodes the input. Args: input (tensor): the input

Get groundtruth from specified gt_sensors. Args: gt_sensors (list): List of strings of sensor names. Valid sensors names: rgb, de

convenience function to see if any files are being loaded Returns: bool: True if loading, False otherwise

Pauses the editor physics simulation

Arguments: in_channels: int, number of channels in the input tensor. Default is 3 for RGB images. dep

Proportion of shared values between two sorted lists at given depth. >>> _round(agreement("abcde", "abdcf", 1)) 1.0 >>> _round(agreement("

Performs an evaluation. Args: val_loader (dataloader): pytorch dataloader

Allow use of omni.kit.commands interface

Generates the output mesh. Args: data (tensor): data tensor return_stats (bool): whether stats should be returned

Makes a 3D grid. Args: bb_min (tuple): bounding box minimum bb_max (tuple): bounding box maximum shape (tuple): output s

Overlap which accounts for possible ties. This isn't mentioned in the paper but should be used in the ``rbo*()`` functions below, otherwise ov

Parse state_dict of model and return scalars. Args: state_dict (dict): State dict of model

Stops the editor physics simulation