torch

hrtfpykit.datasets.torch is the PyTorch integration module for dataset workflows. It keeps PyTorch-specific batching and training-loop APIs separate from the framework-neutral dataset classes, while still giving PyTorch users a stable import path inside hrtfpykit.datasets.

PyTorch is required for this submodule. The core hrtfpykit runtime dependency set leaves PyTorch outside the base installation, so environments that import hrtfpykit.datasets.torch must provide PyTorch themselves.

Import PyTorch integration APIs from this submodule explicitly:

from hrtfpykit.datasets.torch import collate_samples, hrtf_loss

• collate_samples
• hrtf_loss