utils

Modules

`modelopt.torch.utils.cpp_extension`	Utility functions for loading CPP / CUDA extensions.
`modelopt.torch.utils.dataset_utils`	Utility functions for getting samples and forward loop function for different datasets.
`modelopt.torch.utils.distributed`	Utility functions for using torch.distributed.
`modelopt.torch.utils.graph`	Utility functions for computational graph.
`modelopt.torch.utils.image_processor`	Utility classes for image processing.
`modelopt.torch.utils.import_utils`	Handles suppressing import errors for third-party modules that may or may not be available.
`modelopt.torch.utils.list`	Utils for operating on lists.
`modelopt.torch.utils.logging`	Utility functions for logging.
`modelopt.torch.utils.memory_monitor`	GPU Memory Monitoring Utilities.
`modelopt.torch.utils.network`	Utility functions for PyTorch models.
`modelopt.torch.utils.perf`	Utility functions for performance measurement.
`modelopt.torch.utils.random`	Random number generator with a deterministic, synchronized seed for sampling.
`modelopt.torch.utils.tensor`	Utility functions for PyTorch tensors.
`modelopt.torch.utils.vlm_dataset_utils`	Utility functions for getting samples and forward loop function for different vlm datasets.

Utility functions.