Commit Graph

172 Commits

Author SHA1 Message Date
Mike Sperber d088e30b80
Add StatelessCase, TPUCompile to blacklist. (#271)
Needed to support Tensorflow 2.4.

Also, add AttrFunc to unbreak build.
2021-02-09 12:08:46 -05:00
Jiyong Jung b1a8a0513d Updates the build badge path to the new location. 2021-02-08 20:58:37 -05:00
jcmartin 6b19e54722
Update to haddock files for tensorflow-0.3 package (TensorFlow 2.3.0). (#269)
* Update README to refer to 2.3.0-gpu.
* Remove old package documentation from haddock directory.
2020-11-13 12:21:27 -08:00
jcmartin 4d9d315fbd
Update Package Version Numbers (#268) 2020-11-11 09:21:35 -08:00
jcmartin c66c912c32
Tensorflow 2.3.0 Support (#267)
* Tensorflow 2.3.0 building and passing tests.
* Added einsum and test.
* Added ByteString as a possible argument to a function.
* Support more data types for Adam.
* Move to later version of LTS on stackage.
* Added a wrapper module for convolution functions.
* Update ci build to use a later version of stack.
* Removed a deprecated import in GradientTest.
2020-11-06 11:32:21 -08:00
Mike Sperber 568c9b6f03
Update to current proto-lens packages. (#258) 2020-05-21 13:36:52 -07:00
Mike Sperber 0f322b2e06
Fix MonadFail-related errors to support ghc 8.8 2020-04-13 16:48:43 -07:00
Mike Sperber 739f6618f4
Unbreak build with Cabal >= 2.4. (#256)
Remove sDistHook, add autogen-modules.
2020-04-12 10:28:56 -09:00
Saeed Hadikhanloo f07c7f237a
nix-shell updated to be compatible with LTS14.4 (#254) 2020-02-15 21:45:40 -08:00
Greg Steuck 8cde4d6a27
Update stack version in README.md 2019-09-22 22:14:58 -07:00
Greg Steuck 70824616ee
Add autogen-module to upload (related to #249) (#250) 2019-09-08 16:22:06 -07:00
Greg Steuck 580917a731
Switch to LTS 14 and bump versions to 0.2.0.1 (#248)
* Use newer stack and protoc in Dockerfiles
2019-09-08 12:45:24 -07:00
Yorick 7e1d92c6c2 Bump version constraints for proto-lens-* (#247) 2019-09-02 18:19:07 -07:00
Yorick 26eebce98f Upgrade to tensorflow 1.14.0 (#244)
* add curl, unzip, zlib1g-dev to dockerfiles
2019-07-15 23:16:58 +02:00
rschlotterbeck d741c3ee59 Add gradient for batchMatMul (#246) 2019-07-08 13:41:35 -04:00
rschlotterbeck c811037cb9 Add gradient for sigmoid (#245) 2019-07-07 20:18:02 -04:00
Christian Berentsen 1fbd5d41dd Add gradients for DepthwiseConv2dNative (#240) 2019-04-22 00:46:27 -04:00
Christian Berentsen 4a2e46ba57 Make 'mean' doubly differentiable (#241)
Use stopGradient on shape computations
Add opGrad for StopGradient
2019-04-22 00:46:01 -04:00
Daniel YU 7316062c10 upgrade to ghc 8.6.4 (#237) 2019-04-11 19:27:15 -07:00
Christian Berentsen c0f87dc0bc Avoid computing gradients for incidental nodes (#238) 2019-04-11 14:17:19 -04:00
Christian Berentsen 96f1c88327 Add gradient for ResizeBilinear (#239) 2019-04-08 13:43:17 -04:00
erikabor 3cfd96ef08 Add gradient for slice function (#234) 2019-03-26 16:30:50 -04:00
erikabor 666dce94bd Add gradient for sqrt function (#236) 2019-03-18 21:08:08 -04:00
Greg Steuck 896a0d31f7
Added a link to CentOS installation instructions
FYI @subbyte
2019-02-28 19:55:34 -08:00
Rik c7426a3f00 Upgraded Stackage resolver to LTS-12 (GHC 8.4.4) and Nix channel to 18.04 (#231) 2019-02-05 19:32:15 -08:00
Rik e4acd69574 Support gradients of pad, squeeze, spaceToBatchND, and batchToSpaceND (#226) 2018-11-27 14:17:32 -05:00
Rik 95c6b6f277 Added support for ExpandDims gradient. (#224) 2018-11-20 21:45:31 -05:00
Rik 915015018c Added support for tanh activation function (#223) 2018-11-14 12:08:05 -05:00
Christian Berentsen 61e58fd33f Use proto-lens* == 0.3.* (#212)
* Include more *_Fields modules
2018-09-04 10:44:52 -07:00
Greg Steuck 99f8c8cf56
Haddock (#209)
* Symlink the correct directory.

* Update haddock with a symlink
2018-08-17 23:26:23 -07:00
Christian Berentsen c978837cd3 Document use of nvidia docker version 2 (#208) 2018-08-17 16:48:13 -04:00
Matt Ludwigs 8a0be8ebb7 Update README link to point to the most recent release docs (#207) 2018-08-14 20:06:46 -07:00
fkm3 65f374a019 Update checked in haddock (#205) 2018-08-13 21:05:49 -07:00
fkm3 85bf0bb12c
Bump all of the versions to 0.2.0.0 (#202) 2018-08-02 22:07:30 -04:00
Greg Steuck c731a6f768
Merge pull request #201 from helq/master
Update to tensorflow 1.9
2018-07-25 12:19:20 -07:00
Greg Steuck 5be6c2ab5a
Merge pull request #200 from basvandijk/simplified-shell.nix
Simplified shell.nix
2018-07-25 11:10:17 -07:00
helq ecfea97992 Update to tensorflow 1.9 2018-07-24 16:48:55 -05:00
Bas van Dijk ff69c12223 Simplified shell.nix
* I upgraded nixpkgs to the latest master which now contains
  libtensorflow. This replaces the custom tensorflow-c. Note that
  libtensorflow also builds on OS X and has optional CUDA support.

* nixpkgs is now pinned using builtins.fetchTarball which means we
  can build with an empty NIX_PATH thus making the build pure.

  This does require Nix-2.0 but this has been released for a while
  with NixOS-18.03.

* The stack override has been replaced by passing
  `extraArgs = "--system-ghc";` to buildStackProject.
2018-07-24 23:16:45 +02:00
fkm3 e40d2c44f0
Update changelog for v0.2.0.0 (#194) 2018-07-12 15:44:43 -04:00
Nicolas Mattia 8e1d85b5e5 Use stack's nix support (#168)
* Add stack --nix support

* Update README for nix shell
2018-05-24 19:33:15 -07:00
Miki Fossati b063cccc40 Fix install script for High Sierra (#192)
* Copy libtensorflow files from a temporary directory
2018-05-21 08:20:20 -07:00
Erlend Hamberg 88dafe8f3a Update to LTS-11 (#187)
Fix Conduit 1.3 deprecation warnings
- The `Conduit`, `Producer` and `Consumer` aliases are deprecated and
  `ConduitT` is used directly instead
- `=$=` is deprecated and replaced by `.|`
2018-05-18 19:20:11 -04:00
fkm3 baa501b262
Use newer version of stack in CI (#189)
Required by #187.

The version we were using is old enough that it doesn't work with the
latest stackage LTS. haskellstack.org says

    There is also a Ubuntu package for Ubuntu 16.10 and up, but the
    distribution's Stack version lags behind, ...

So, instead of asking them to update it, it's probably better to
download the tar of the version we want.

Somehow updating stack surfaced a new pedantic warning in GradientTest,
so I've fixed that as well.
2018-05-15 23:19:15 -04:00
Erlend Hamberg 1829c7ef3c Update to tensorflow 1.8 (#188) 2018-05-14 15:55:33 -04:00
fkm3 1e2dca8701
Update to tensorflow 1.7 (#185)
All of the non-s/1.3/1.7/ changes are because

* There are new tensorflow datatypes
* Some ops have looser types (e.g. fill now accepts both int64 and int32)
* There are more ops of type "func"
2018-04-17 12:24:31 -04:00
fkm3 e35211d49b Fix initialized variables for tensorflow 1.7 (#184)
* Fix initialized variables for tensorflow 1.7

This is needed to support tensorflow 1.7. The trick of initializing a
variable with `Shape []` and then overriding the shape by assigning an
initial value no longer works. It seems that we need to explicitly flip
the unknown_rank bit in the shape proto.

I thought about switching opgen to use `Maybe Shape` when an op requires
a shape attribute, but that will cause a lot of api churn, so I chose to
hold off for now and just do a spot fix to unblock 1.7.
2018-04-16 07:48:05 -07:00
Noon van der Silk 4c6306d914 fixes #182 (#183) 2018-04-10 21:12:48 -04:00
fkm3 96c3025171
Fix CI build badge (#181) 2018-04-10 19:32:53 -04:00
fkm3 3cd5184716
Update README.md 2018-04-09 16:29:32 -04:00
Christian Berentsen f2cafa7071 Added imageSummary wrapper (#159) 2018-01-23 13:02:58 -05:00