Skip to content

Releases: torch-points3d/torch-points-kernels

0.7.0

16 Mar 09:03
db05fae
Compare
Choose a tag to compare

Change

  • Added some extra compilation flags: FORCE_CUDA=1 for CUDA install and FORCE_ONLY_CPU=1 for CPU only install.

Grid sampling transform for dense data

23 Dec 09:35
88dfbd5
Compare
Choose a tag to compare
v0.6.10

Update setup.py

Cubic feature sampling

18 Dec 10:31
a52ea03
Compare
Choose a tag to compare

Thank you @hzxie for the addition!

Bug fix

03 Aug 08:39
e99099b
Compare
Choose a tag to compare

0.6.8

Bug fix

  • Comilation with older GPU architecture

Chamfer distance

08 Jul 13:37
eb64e4c
Compare
Choose a tag to compare

Additions

Maintainance

08 Jul 12:36
83ac174
Compare
Choose a tag to compare

Additions

  • Windows support

Change

  • Develop with python 3.7

Bug fix

  • Fixed bug in region growing related to batching
  • Ball query for partial dense data on GPU was returning only the first point. Fixed now

Adds panoptic support

30 Jun 12:24
0220fa5
Compare
Choose a tag to compare

Additions

  • Clustering algorithm for PointGroup
  • Instance IoU computation on CPU and GPU

Change

  • Force no ninja for the compilation

0.6.4

22 May 10:36
8cc7b07
Compare
Choose a tag to compare

MacOS and Windows support

0.6.3

20 May 10:57
16e2846
Compare
Choose a tag to compare

Removed pyproject.toml to avoid PEP-517 issues

0.6.2

20 May 07:51
d5c27f6
Compare
Choose a tag to compare
  • Fix install with pip > 19