mirror of
https://github.com/tensorflow/haskell.git
synced 2024-06-28 15:38:34 +02:00
d62c614695
Distinguish between "rendered" and "unrendered" Tensors.
There are now three types of `Tensor`:
- `Tensor Value a`: rendered value
- `Tensor Ref a`: rendered reference
- `Tensor Build a` : unrendered value
The extra bookkeeping makes it easier to track (and enforce) which tensors are
rendered or not. For examples where this has been confusing in the past, see
With this change, pure ops look similar to before, returning `Tensor Build`
instead of `Tensor Value`. "Stateful" (monadic) ops are unchanged. For
example:
add :: OneOf [..] t => Tensor v'1 t -> Tensor v'2 t -> Tensor Build t
assign :: (MonadBuild m, TensorType t)
=> Tensor Ref t -> Tensor v'2 t -> m (Tensor Ref t)
The `gradients` function now requires that the variables over which it's
differentiating are pre-rendered:
gradients :: (..., Rendered v2) => Tensor v1 a -> [Tensor v2 a]
-> m [Tensor Value a]
(`Rendered v2` means that `v2` is either a `Ref` or a `Value`.)
Additionally, the implementation of `gradients` now takes care to render every
intermediate value when performing the reverse accumulation. I suspect this
fixes an exponential blowup for complicated expressions.
53 lines
1.9 KiB
Haskell
53 lines
1.9 KiB
Haskell
-- Copyright 2016 TensorFlow authors.
|
|
--
|
|
-- Licensed under the Apache License, Version 2.0 (the "License");
|
|
-- you may not use this file except in compliance with the License.
|
|
-- You may obtain a copy of the License at
|
|
--
|
|
-- http://www.apache.org/licenses/LICENSE-2.0
|
|
--
|
|
-- Unless required by applicable law or agreed to in writing, software
|
|
-- distributed under the License is distributed on an "AS IS" BASIS,
|
|
-- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
-- See the License for the specific language governing permissions and
|
|
-- limitations under the License.
|
|
|
|
{-# LANGUAGE OverloadedLists #-}
|
|
module Main where
|
|
|
|
import Control.Monad.IO.Class (liftIO)
|
|
import Data.Int (Int64)
|
|
import Google.Test (googleTest)
|
|
import Test.Framework (Test)
|
|
import Test.Framework.Providers.HUnit (testCase)
|
|
import Test.HUnit ((@=?))
|
|
import qualified Data.Vector as V
|
|
|
|
import qualified TensorFlow.Ops as TF
|
|
import qualified TensorFlow.Core as TF
|
|
import qualified TensorFlow.GenOps.Core as CoreOps
|
|
|
|
-- | Test split and concat are inverses.
|
|
testSplit :: Test
|
|
testSplit = testCase "testSplit" $ TF.runSession $ do
|
|
let original = TF.constant [2, 3] [0..5 :: Float]
|
|
splitList = CoreOps.split 3 dim original
|
|
restored = CoreOps.concat dim splitList
|
|
dim = 1 -- dimension to split along (with size of 3 in original)
|
|
liftIO $ 3 @=? length splitList
|
|
(x, y, z) <- TF.run (original, restored, splitList !! 1)
|
|
liftIO $ x @=? (y :: V.Vector Float)
|
|
liftIO $ V.fromList [1, 4] @=? z
|
|
|
|
testShapeN :: Test
|
|
testShapeN = testCase "testShapeN" $ TF.runSession $ do
|
|
let shapes = map TF.Shape [[1],[2,3]]
|
|
let tensors = map TF.zeros shapes :: [TF.Tensor TF.Build Float]
|
|
result <- TF.run $ CoreOps.shapeN tensors
|
|
liftIO $ [V.fromList [1], V.fromList [2,3]] @=? (result :: [V.Vector Int64])
|
|
|
|
main :: IO ()
|
|
main = googleTest [ testSplit
|
|
, testShapeN
|
|
]
|