Overview

Data keys and collation, conversions, the Hub token, optional imports, I/O, random state, checkpoints, and types.

Modules:

• conversion –

  Conversions between iterables, batch index representations (offset, bincount, cu_seqlens), and tensor formats.

• data –

  Data loading: standard sample keys, packed-batch collation and row offsets, worker seeding, and the data loader.

• hub –

  Hugging Face Hub file URLs and the token of the environment.

• imports –

  Optional dependency detection, lazy imports, and availability flags.

• io –

  File reading and writing for point cloud data: JSON, OFF meshes, PLY, and safetensors.

• misc –

  Miscellaneous helpers for parallel mapping and nested attribute access.

• random –

  Random seeding and determinism control.

• state_dict –

  Checkpoint reading, adaptive state_dict loading, and key-mapping transforms.

• types –

  Shared type aliases, TypedDict definitions, and enums used across the package.