r"""Return dataset object for the test set Args: cfg (obj): Global configuration file. subset_indices (sequence): Indices of the

r"""Return dataset objects for the training set. Args: cfg (obj): Global configuration file. subset_indices (sequence): Indices of

r"""Return dataset objects for the validation set. Args: cfg (obj): Global configuration file. subset_indices (sequence): Indices

r"""Every trainer should implement its own init loss function.

Prints the profiling results and resets the timers.

r"""Operations to do before starting an epoch. Args: current_epoch (int): Current number of epoch.

r"""Operations to do before starting an iteration. Args: data (dict): Data used for the current iteration. current_it

r"""Record time before applying backward.

r"""Record time before applying forward.

r"""Record forward, backward, loss, and model average time for the network update.

r"""Record time before computing loss.

r"""Record time before applying model average.

r"""Record time before updating the weights

r"""Wrap a model for distributed data parallel training. Args: model (obj): PyTorch network model. Returns: (obj): Wrapped P

r"""Broadcast object list from the master to the others

Aggregate the list of test data from all devices and process the results. Args: data_batches (list): List of (hierarchical) dictionaries,

Contract function in mip-NeRF 360 (eq 10). Args: x (tensor [...,3]): The input points. Returns: x_warp (tensor [...,3]): The

Jacobian of the contract function in mip-NeRF 360. Args: x (tensor [...,3]): The input points. Returns: jacobian (tensor [...

Matrix form of cross product opertaion. param x: [3,] tensor. return: [3, 3] tensor representing the matrix form of cross product.

r""" gather to all ranks

r""" Reduce to all ranks

r"""Things to do after an epoch. Args: data (dict): Data used for the current iteration. current_epoch (int): Curren

Compute Essential Matrix from a relative pose. param poses: [views, 3, 4] tensor representing relative poses. return: [views, 3, 3] tensor re

r"""Get CPU affinity

Generate a unique id based on date. Returns: str: Return uid string, e.g. '20171122171307111552'.

r"""Return the optimizer object. Args: cfg_optim (obj): Config for the specific optimization module (gen/dis). model (obj): PyTor

Allow the network to use different hyperparameters (e.g., learning rate) for different parameters. Returns: PyTorch parameter grou

Return the scheduler object. Args: cfg_opt (obj): Config for the specific optimization module (gen/dis). opt (obj): PyTorch optim

Return the scheduler object. Args: cfg_optim (obj): Config for the specific optimization module (gen/dis). optim (obj): PyTorch o

r"""Return dataset objects for testing Args: cfg (obj): Global configuration file. subset_indices (sequence): Indices of the subs

r"""Return dataset objects for the training and validation sets. Args: cfg (obj): Global configuration file. shuffle (bool): Wheth

Aggregate the list of test data from all devices and process the results. Args: data_gather (dict): (hierarchical) dictionaries, where lea

r"""Return dataset objects for the training and validation sets. Args: cfg (obj): Global configuration file. subset_indices (seque

r"""Initialize automatic mixed precision training.

r"""Initialize the cudnn module. The two things to consider is whether to use cudnn benchmark and whether to use cudnn deterministic. If cudnn

r"""Create log directory for storing checkpoints and output images. Args: config_path (str): Path to the configuration file. logd

r"""Initialize logging attributes.

r"""Initialize loss functions. All loss names have weights. Some have criterion modules.

r"""Initialize validation parameters.

r"""Initialize Weights & Biases (wandb) logger. Args: cfg (obj): Global configuration. wandb_id (str): A unique ID fo