Dict as attribute trick.

r"""Configuration class. This should include every human specifiable hyperparameter values for your training.

A class of operations on camera poses (PyTorch tensors with shape [...,3,4]). Each [3,4] camera pose takes the form of [R|t].

r"""Device used for nvml.

r"""Sampler that supports loading from an iteration. This is very useful for preemptable jobs. Args: dataset (torch.utils.data.Datase

Lie algebra for SO(3) and SE(3) operations in PyTorch.

r"""In this model average implementation, the spectral layers are absorbed in the model parameter by default. If such options are turned on, b

r"""Dummy wrapping the module.

Sampler that repeats forever. Args: sampler (Sampler)

r"""Base class for image/video datasets. Args: cfg (Config object): Input config. is_inference (bool): Training if False, else va

A customized BaseTrainer.

r"""Base trainer. We expect that all trainers inherit this class. Args: cfg (obj): Global configuration. is_inference (bool): if

Meter is to keep track of statistics along steps. Meters write values for purpose like printing average values. Meters can be flushed to log f

Relentlessly sample from the dataset. This eliminates the overhead of prefetching data before each epoch. https://github.com/rwightman/py