Torchlm Versions Save

💎A high level pipeline for face landmarks detection, it supports training, evaluating, exporting, inference(Python/C++) and 100+ data augmentations, can easily install via pip.

torchlm-0.1.6-alpha

2 years ago

torchlm-v0.1.5

2 years ago

v0.1.5 release notes

  • add data augmentations api docs
  • some bugs fixed
  • add more examples

torchlm-v0.1.4

2 years ago

v0.1.4 release notes

  • add _transforms_api_assert
  • fixed apply_background
  • add MANFEST.in
  • add MixUp
  • package transforms assets in to pypi for MixUp

torchlm-v0.1.3

2 years ago

v0.1.3 release notes

  • fixed wrong flag setting when the landmarks been outside
  • fixed docs typos
  • inti metrics module

torchlm-v0.1.2

2 years ago

v0.1.2 release notes

  • add safe check and make sure the landmarks keep the same number after any transforms
  • fixed docs typos error
  • add internal transform logging api
  • add prob param at bind-level

torchlm-v0.1.1

2 years ago

🆕 What's New

  • [2022/02/13]: Add 30+ native data augmentations and bind 80+ torchvision and albumations's transforms.

torchlm-v0.1.0

2 years ago

🆕 What's New

  • [2022/02/13]: Add 30+ native data augmentations and bind 80+ torchvision and albumations's transforms.