unipose API Documentation Modules Classes Names
Clear Help

Search bar offers the following options:

  • Term presence. The below example searches for documents that must contain “foo”, might contain “bar” and must not contain “baz”: +foo bar -baz
  • Wildcards. The below example searches for documents with words beginning with “foo”: foo*
  • Search in specific fields. The following search matches all objects in "twisted.mail" that matches “search”: +qname:twisted.mail.* +search

    Possible fields: 'name', 'qname' (fully qualified name), 'docstring', and 'kind'. Last two fields are only applicable if "search in docstrings" is enabled.

  • Fuzzy matches. The following search matches all documents that have a word within 1 edit distance of “foo”: foo~1

Results provided by Lunr.js

Module Index

  • unipose - No package docstring; 3/4 packages documented
    • datasets - Datasets for pose estimation. All datasets are subclasses of unipose.dataset.BaseJointDataset.
      • animal_kingdom - Undocumented
      • base_joint_dataset - Base class for datasets that have keypoints and images.
      • coco - Undocumented
      • mpii - Undocumented
      • multi_dataloader - Undocumented
      • utils - No module docstring; 5/7 functions documented
    • layers - Undocumented
      • bottleneckx - Undocumented
      • duc - Undocumented
      • selayer - Undocumented
      • seresnext - Undocumented
    • losses - Skeleton consistency and graph learning losses.
      • graph_learning - Graph Learning by Smoothness (GLES) Loss
    • models - Implements UniPose model.
      • unipose - Undocumented
API Documentation for unipose, generated by pydoctor 22.9.1 at 2023-01-06 17:14:35.