Skip to content
GitLab
Explore
Sign in
Maksymilian Grochowski
214061
Repository
Branches
Overview
Active
Stale
All
release_v0.1
d99027d6
·
Update nvJPEG to 0.1.2 (#79) (#86)
·
Jul 30, 2018
release_v0.2
2319d246
·
Add int64 as label type and set it as default (#125)
·
Aug 21, 2018
release_v0.3
74d9b580
·
Use pinned memory only when interfacing with MakeContiguous
·
Sep 17, 2018
release_v0.4
2a5e14c6
·
Remove double-buffering on the MXNet side (#258)
·
Oct 31, 2018
release_v0.5
235ae804
·
Partial cherry-pic of #305 (#307)
·
Nov 16, 2018
release_v0.6
c8d9d725
·
Change nvcuvid link to dyn + hint for FFmpeg (#370)
·
Dec 15, 2018
release_v0.6.1
d5e1c73b
·
Add fallback build configuration in case of no CUDA runtime library a… (#427)
·
Jan 14, 2019
release_v0.7
51804e9c
·
Change version to 0.7.0
·
Feb 05, 2019
release_v0.8
ad8699d8
·
Stop using tf-nightly-gpu since it broke the build. Use latest released version instead (#730)
·
Apr 09, 2019
release_v0.9
666ce55e
·
Fix initialization of CUDA context on the default device during pipeline creation (#829)
·
Apr 28, 2019
release_v0.10
527e2fd0
·
Better PythonFunction error message (#883)
·
May 16, 2019
release_v0.11
71c499c5
·
Add as_tensor with provided shape method to python API (#953)
·
Jun 14, 2019
release_v0.12
b9d05de5
·
Reduced batch size to avoid out of memory condition in 19.07 container. (#1089)
·
Jul 19, 2019
release_v0.13
d1e100e5
·
Fix fp16 bug from #1129 and add fp16 test case (#1160)
·
Aug 12, 2019
release_v0.14
983b66d3
·
Set version to 0.14.0
·
Sep 06, 2019
release_v0.15
67744062
·
Relax check for libnvidia-opticalflow is test script. (#1381)
·
Oct 16, 2019
release_v0.16
b33c5ace
·
Fix scalar batch handling in arithmetic ops (#1449)
·
Nov 08, 2019
release_v0.17
e61c304d
·
Fix problem with TensorFlow and cupy tests (#1568)
·
Dec 12, 2019
release_v0.18
e10a3655
·
Update the version to 0.18.0
·
Jan 08, 2020
release_v0.19
6d532d49
·
Change version to 0.19.0
·
Feb 10, 2020
Prev
1
2
3
Next