Commit Graph

2220 Commits

Author SHA1 Message Date
Manjunath Bhat
747e01ea02
Test to check for spurious promotions 2020-01-13 18:33:30 +05:30
Manjunath Bhat
8a93be8c6c
Change loss to cost 2019-12-09 20:39:46 +05:30
thebhatman
d591b2b59e Removed colon and capitalised 2019-10-09 21:36:40 +05:30
thebhatman
96a23c295c Changes to docs 2019-10-09 14:53:03 +05:30
thebhatman
ec886c8ce8 Added docstring for hinge loss 2019-10-03 21:13:09 +05:30
Manjunath Bhat
2b30319a55
Merge branch 'master' into patch-6 2019-09-30 21:05:02 +05:30
thebhatman
ec35e9cbaa Loss functions docs added in layers.md 2019-09-30 21:02:13 +05:30
thebhatman
6e289ef939 Merge branch 'patch-6' of https://github.com/thebhatman/Flux.jl into patch-6 2019-09-30 20:55:44 +05:30
bors[bot]
e2b93bc78a
Merge #874
874: Move CUDNN wrappers to CuArrays r=MikeInnes a=MikeInnes



Co-authored-by: Tim Besard <tim.besard@gmail.com>
Co-authored-by: Mike Innes <mike.j.innes@gmail.com>
2019-09-27 14:05:37 +00:00
Mike Innes
b90b02872f Merge branch 'master' into tb/cuarrays_dnn 2019-09-27 14:58:32 +01:00
Mike Innes
e287982b78 use CuArrays master 2019-09-27 14:55:30 +01:00
Mike Innes
691a29cf32 cudnn bug is fixed 2019-09-27 14:15:58 +01:00
Mike Innes
46bc8e5e64 move pullbacks to CuArrays 2019-09-26 17:14:18 +01:00
bors[bot]
12bc06136d
Merge #870
870: Fix printing of SkipConnection r=MikeInnes a=mcabbott

Before:
```
julia> SkipConnection(Dense(2,2),+)
SkipConnection(Error showing value of type SkipConnection:
ERROR: MethodError: no method matching iterate(::Dense{typeof(identity),TrackedArray{…,Array{Float32,2}},TrackedArray{…,Array{Float32,1}}})

julia> SkipConnection(Chain(Dense(2,3), Dense(3,2), LayerNorm(2)),+)
SkipConnection(Dense(2, 3), Dense(3, 2), LayerNorm(2))

julia> SkipConnection(Dense(2, 3), Dense(3, 2), LayerNorm(2))
ERROR: MethodError: no method matching SkipConnection(::Dense{typeof(identity),TrackedArray{…,Array{Float32,2}},TrackedArray{…,Array{Float32,1}}}, ::Dense{typeof(identity),TrackedArray{…,Array{Float32,2}},TrackedArray{…,Array{Float32,1}}}, ::LayerNorm{TrackedArray{…,Array{Float32,1}}})
```
After:
```
julia> SkipConnection(Dense(2,2),+)
SkipConnection(Dense(2, 2), +)

julia> SkipConnection(Chain(Dense(2,3), Dense(3,2), LayerNorm(2)),+)
SkipConnection(Chain(Dense(2, 3), Dense(3, 2), LayerNorm(2)), +)

julia> SkipConnection(Dense(2,2), (a,b) -> a .+ b./2)
SkipConnection(Dense(2, 2), #9)
```

Co-authored-by: Michael Abbott <32575566+mcabbott@users.noreply.github.com>
2019-09-25 14:09:28 +00:00
Michael Abbott
806e0c5c57 line 2019-09-25 15:20:13 +02:00
Michael Abbott
4245d9acad eg 2019-09-25 15:18:40 +02:00
Michael Abbott
2de84ce79f simplify 2019-09-25 13:59:32 +02:00
Michael Abbott
1a1a96571a +Chain 2019-09-25 13:47:29 +02:00
Michael Abbott
19830c71b1 fix printing of SkipConnection 2019-09-25 13:37:01 +02:00
bors[bot]
acb6a89245
Merge #865
865: Functor r=MikeInnes a=MikeInnes

This refactors our current `@treelike` infrastructure. It somewhat formalises what we're doing around the idea of a Flux model as a functor, i.e. something that can be mapped over.

This is much more flexible than what we had before, and avoids some issues. It allows layers to have state that isn't mappable; it allows for dispatch when walking the tree, which means layers like `BatchNorm` can have non-trainable parameters; and it also allows for zipped mapping like `fmap(+, xs, ys)`, which isn't implemented yet but will be useful for the new optimisers work.

The main downside is that the term `functor` has been previously used in the Julia community as a malapropism for "thing that behaves like a function"; but hopefully this can start to reduce that usage.

Co-authored-by: Mike Innes <mike.j.innes@gmail.com>
2019-09-24 16:36:10 +00:00
bors[bot]
d57636fd48
Merge #861
861: GPU CI maintainance  r=dhairyagandhi96 a=dhairyagandhi96



Co-authored-by: Dhairya Gandhi <dhairya@juliacopmuting.com>
2019-09-24 16:06:13 +00:00
Dhairya Gandhi
ce910da948 compat julia v1.0 2019-09-24 17:04:13 +05:30
Dhairya Gandhi
cf593a5744 revert to custom target 2019-09-24 16:43:48 +05:30
Dhairya Gandhi
fe4ecc5880 trying out extending directly 2019-09-24 16:15:48 +05:30
Dhairya Gandhi
928b5dcc2a fix Zygote 2019-09-24 00:51:35 +05:30
Dhairya Gandhi
822288d63d merge conflicts 2019-09-24 00:31:44 +05:30
Dhairya Gandhi
d8a069b304 fix env 2019-09-24 00:28:52 +05:30
Dhairya Gandhi
98308a85ea add gitlab common yaml 2019-09-23 16:55:53 +05:30
Dhairya Gandhi
783ae137e1 remove targets and env 2019-09-23 16:51:11 +05:30
Dhairya Gandhi
6846551f57 fix cuda init 2019-09-22 22:02:05 +05:30
Dhairya Gandhi
787097f9ea use CuArrays#stable 2019-09-21 00:20:54 +05:30
Mike Innes
b60df53ba1 pkg up 2019-09-19 18:33:33 +01:00
Mike Innes
cabb81e30b internal rename 2019-09-19 15:53:31 +01:00
Mike Innes
b951377426 fix normalisation layer params 2019-09-19 15:33:24 +01:00
Mike Innes
6529dbcbe6 functor refactor 2019-09-19 15:22:11 +01:00
Mike Innes
2c71fc282b rename functor.jl 2019-09-19 14:15:28 +01:00
Mike Innes
f8d5d3b5fc broken normalisation layer params 2019-09-19 14:12:11 +01:00
Dhairya Gandhi
99b6fe57e9 extend test template 2019-09-18 12:32:11 +05:30
Dhairya Gandhi
37fe91d54d remove branch restrictions 2019-09-18 12:05:31 +05:30
Mike Innes
c5e56b7e04 move setweights and copy_transpose 2019-09-17 17:22:35 +01:00
Mike Innes
5baebf48f4 Merge branch 'master' into tb/cuarrays_dnn 2019-09-17 16:17:09 +01:00
Mike Innes
fc9db7ee74 pkg up 2019-09-17 15:49:48 +01:00
Mike Innes
368b1f53b4 tuple support 2019-09-17 15:49:39 +01:00
Mike Innes
b348b20452 cudnn rnns + implicit gradients 2019-09-17 15:41:42 +01:00
Mike Innes
fe57215b7e test fillarray gradients 2019-09-17 15:21:03 +01:00
Dhairya Gandhi
29eae312b8
Merge pull request #863 from Naba7/fix_typo
removed extra parenthesis
2019-09-14 11:43:20 +05:30
Naba7
a600a9ceed removed extra parenthesis 2019-09-14 10:56:17 +05:30
Tim Besard
6ea2557c46 Use correct CuArrays branch for CI. 2019-09-13 08:21:45 +02:00
Tim Besard
4942d7fcfd Move functionality over to CuArrays. 2019-09-13 08:21:45 +02:00
Tim Besard
1e7ff4f65d Query the worksize. 2019-09-13 08:04:05 +02:00