1
0
Fork 0
mirror of https://github.com/tensorflow/haskell.git synced 2024-11-23 03:19:44 +01:00
Commit graph

157 commits

Author SHA1 Message Date
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
fkm3
760c067e89
Add tensorflow-haskell-deptyped example (#174)
https://github.com/tensorflow/haskell/issues/156
2018-01-17 13:31:23 -05:00
fkm3
bc15709cb4
Update outer_launch_tests.sh to support new internal CI (#173) 2018-01-16 16:16:41 -05:00
fkm3
e4b4c344ab Add a link to the project using dependent types for more type safety. 2017-12-21 17:24:19 -08:00
Chris Roth
cbd607f4d6 Update Link to TensorFlow.Core (#167) 2017-11-11 17:01:52 -05:00
Benjamin
0391c2dbb7 Change MAINTAINER to LABEL since the former is deprecated (#164) 2017-10-24 19:31:13 -07:00
fkm3
a8405d15a2 Use newer tensorflow image in ci_build/Docker (#163)
Not sure if this makes any difference, but it seems like a good idea for consistency.
2017-10-21 21:24:40 -07:00
Christian Berentsen
5c11ce1f03 How to use GPU with stack, docker, nvidia-docker (#157) 2017-10-21 20:32:19 -04:00
fkm3
fe64a72f42 Update haddock (#162) 2017-10-19 20:56:38 -07:00
fkm3
7720af0afd Update to tensorflow 1.3 (#161)
* Tested on linux without Docker.
* Couldn't get nix build to work, so I just updated the URL and hash.
* Did not test macos build.

The mnist change was necessary because the argmax output type is now polmorphic.
2017-10-19 13:41:55 -04:00
Christian Berentsen
2dcc921f6e Gradient of Conv2DBackpropInput (#155) 2017-10-15 11:49:44 -07:00
Jeroen Bransen
d8bf349962 Create monad transformer version of Session (closes #153) (#154) 2017-10-02 16:33:49 -04:00
Nathan T.A. Lewis
d79a919efa Fix a small typo in the warning message of numOutputs (#151) 2017-08-24 17:34:22 -04:00
Alan Yee
7e0ef4bcd3 Update README.md (#149)
Add license
2017-08-21 15:59:22 -04:00
Judah Jacobson
7328cb277f Fix the build with ghc-8.2.1. (#147)
- Avoid using a deprecated Cabal function
- Use newer versions of proto-lens packages in stack.yaml
- Work around a new type-level warning that affects `OneOf/TensorTypes`.
2017-08-08 09:48:59 -07:00
Judah Jacobson
56038ba27e Depend on a newer proto-lens and remove orphan Ord instances. (#146)
proto-lens-0.2.2.0 generates Ord instances for all message types,
so we can remove the orphan instances we previously added.

Dependends on proto-lens-protoc-0.2.2.1 or newer due to google/proto-lens#113.
2017-08-02 22:47:55 -07:00