062a7aaf33
Build against TF 2.15 ( #2857 )
...
* Built against TF2.15
2023-11-27 17:10:27 -08:00
48f1125e6c
Build with TF 2.14 ( #2856 )
...
* Built with tf2.14 patch
2023-11-19 20:35:03 -08:00
9c1642aac7
Build 2.12 ( #2821 )
...
* Initial build against TF2.12
* support 3.11
---------
Co-authored-by: bhack <bhack@users.noreply.github.com>
2023-04-07 07:22:53 -07:00
c80423ad8b
Build with TF 2.11.0 ( #2782 )
...
* Build for TF 2.11.0
Co-authored-by: Sean Morgan <seanmorgan@outlook.com>
2022-12-09 13:21:42 -08:00
2e772fd6b0
Build Against TF 2.10 ( #2744 )
...
* Build TF 2.10rc0
* Build with C++17
Co-authored-by: bhack <bhack@users.noreply.github.com>
Co-authored-by: Lukas Geiger <lukas.geiger94@gmail.com>
2022-09-20 16:27:44 -07:00
1c3a2b2b0c
Build against TF 2.9 ( #2696 )
...
* Build with TF 2.9
* Update cuda toolchain for manylinux2014
2022-05-20 14:21:30 -07:00
8cc368e7c2
Build Addons for Windows (CPU) ( #771 )
...
* Update bazel build to compile with MSVC
* Update custom ops to use C++ standard
* Remove Windows ParseTime until TF core exports symbols
* Fix resource loader on windows
* Disable parse time on windows
* Fix packaging
* Finalize windows build
2019-12-19 12:43:28 -05:00
6fd4580281
Build using CUDA 10.1 ( #730 )
...
* Build using bazel > 1.0 and CUDA 10.1
* Fix toolchain name
* Update bazel version for macos build
* Remove depset iteration for bazel 1.x+
* Working 10.1 build
* Run bazel tests for single python environment.
Otherwise in bazel > 1.0 we would need to install tf-nightly on py2 and
py3 in order to test.
* Make CI testing choose default python on path
* Touch up README
2019-12-04 11:19:06 -08:00
d20e803f26
BLD: Fix default GPU toolchain ( #524 )
...
* Refactor build to work outside of docker
2019-10-01 09:09:11 -04:00
8309529d9d
Add correlation cost layer ( #207 )
...
* Add correlation cost layer
2019-07-30 13:35:07 -04:00
00cc3b6f49
BLD: Restructure directories ( #356 )
2019-07-15 22:55:48 -04:00
a7afaa703e
set up GPU environment ( #294 )
...
* BLD: compile gpu kernel for image_projective_transform_op_gpu
* BUG: minor fix
* TST: use gpu
* BLD: fix code style
* CLN: move gpu dir to external dir
* CLN: configure ABI flag
* CLN: leave TODO in configure.sh
* CLN: create cuda_libs target
* CLN: fork use_gpu
* CLN: fix code style
* CLN: remove -c opt and use CUDA_HOME
* CLN: rename external => build_deps
* CLN: add requirements_gpu.txt
* TST: disable failed test cases on GPU
* CLN: tracked by compile issue
2019-07-12 20:39:24 -04:00
199432bf40
Update tf dependency folder
2019-01-10 10:30:20 -05:00
2b004243f5
Directory structure for discussion
2019-01-04 14:06:31 -05:00
Sean Morgan