Commit Graph

2305 Commits

Author SHA1 Message Date
Mike Innes
b65b491e51 compat, pkg up 2019-11-28 16:23:22 +00:00
Mike J Innes
75d609ecc8
Update README.md 2019-11-28 16:00:55 +00:00
Mike J Innes
99f98ca800
Update README.md 2019-11-28 16:00:21 +00:00
Tim Besard
ab450477f3
Merge pull request #944 from FluxML/rnn-fix
RNN failure hackaround
2019-11-27 16:06:13 +01:00
Mike Innes
1c0e9acc45 Update CuArrays to include the workspace fix. 2019-11-27 14:31:03 +01:00
bors[bot]
90a38a3201
Merge #937
937: Fix Glorot initialization, add He initialization r=MikeInnes a=Sleort

Should fix #442 .
Adds He weight initialization as a bonus :-)

Co-authored-by: Troels Arnfred Bojesen <tr-ab@online.no>
2019-11-26 16:17:06 +00:00
bors[bot]
fb4a48f970
Merge #943
943: Fixes #900 r=MikeInnes a=dhairyagandhi96

Thoughts on the test?

cc @MikeInnes

Co-authored-by: Dhairya Gandhi <dhairya@juliacopmuting.com>
2019-11-26 15:09:27 +00:00
Dhairya Gandhi
59bb0d81b0 add TODO 2019-11-26 16:23:09 +05:30
Mike J Innes
4c69b44a7c
Merge pull request #940 from matsueushi/feature/cuda-logitbc
Fix logitbinarycrossentropy on CuArrays
2019-11-26 10:18:07 +00:00
Dhairya Gandhi
c031ae1a94 correct channel value 2019-11-24 13:31:31 +05:30
Tim Besard
fbb377a7b4
Merge pull request #941 from FluxML/tb/include_during_precompile
Don't include the CUDA module during precompilation.
2019-11-24 08:55:43 +01:00
Dhairya Gandhi
5f21238d1a no grad dims helper 2019-11-24 13:25:02 +05:30
Tim Besard
4ece13c649 Don't include the CUDA module during precompilation.
If we do, we could end up replacing it at runtime.
2019-11-22 18:03:51 +01:00
matsueushi
a0314ce682 Fix logitbinarycrossentropy on CuArrays 2019-11-22 05:23:24 +00:00
Troels Arnfred Bojesen
3f97701d4c Merge branch 'HEAD' into weight_init_patch 2019-11-20 13:25:32 +09:00
Troels Arnfred Bojesen
60a29abaf1 Merge branch 'weight_init_patch' into HEAD 2019-11-20 13:25:19 +09:00
Troels Arnfred Bojesen
3b83828e4e Merge branch 'HEAD' into weight_init_patch 2019-11-20 13:24:48 +09:00
Troels Arnfred Bojesen
af96a197c1 Fix Glorot initialization
Should fix #442
2019-11-20 13:20:42 +09:00
Mike J Innes
5839e166f6
Merge pull request #860 from dsweber2/activations
Activations
2019-11-19 16:44:25 +00:00
Tim Besard
2fa3e5673e
Merge pull request #924 from FluxML/tb/cuda_init
CUDA package initialization improvements
2019-11-19 16:48:45 +01:00
Tim Besard
c45cec4cba Simplify warning. 2019-11-19 16:05:41 +01:00
Tim Besard
bd734ed957 Bump CUDA dependencies. 2019-11-19 15:55:25 +01:00
Tim Besard
69bf84278f Remove wrong warning. 2019-11-19 15:53:43 +01:00
Mike J Innes
4f73e434a4
Merge pull request #935 from baggepinnen/patch-4
Fix AMSGrad on GPU
2019-11-19 12:58:37 +00:00
Troels Arnfred Bojesen
2b80573248 Fix Glorot initialization, add He initialization
Should fix #442 .
Adds He weight initialization as a bonus :-)
2019-11-19 18:16:29 +09:00
bors[bot]
8638bcdcd7
Merge #936
936: Avoid unnecessary conversion r=MikeInnes a=baggepinnen

This initialization works for both cpu and gpu

Co-authored-by: Fredrik Bagge Carlson <baggepinnen@gmail.com>
2019-11-19 09:05:23 +00:00
Fredrik Bagge Carlson
2da22f31f0
Avoid unnecessary conversion
This initialization works for both cpu and gpu
2019-11-19 16:31:04 +08:00
Fredrik Bagge Carlson
df7ffb0ef8
Fix AMSGrad on GPU
The previous initialization created a CPU array. Now, the same type of array as `x` is created.
2019-11-19 16:27:44 +08:00
Troels Arnfred Bojesen
4530ac65c7 Fix Glorot initialization, add He initialization
Should fix the issue reported at https://github.com/FluxML/Flux.jl/issues/442 .
Adds He weight initialization as a bonus :-)
2019-11-19 16:50:40 +09:00
Mike J Innes
967cc1c175
Merge pull request #927 from heliosdrm/patch-1
Extend docs about `train!`
2019-11-18 12:22:16 +00:00
dsweber2
dea29532ef Merge branch 'master' into activations 2019-11-15 17:19:43 -08:00
Helios De Rosario
a0e3729679
Update docs/src/training/training.md
Co-Authored-By: Mike J Innes <mike.j.innes@gmail.com>
2019-11-15 21:17:45 +01:00
dsweber2
20eb840882 keeping activations separate 2019-11-15 12:03:08 -08:00
bors[bot]
7eb6a0c98c
Merge #932
932: Travis: test on 1.0 r=MikeInnes a=MikeInnes



Co-authored-by: Mike J Innes <mike.j.innes@gmail.com>
Co-authored-by: Mike Innes <mike.j.innes@gmail.com>
2019-11-15 16:21:30 +00:00
Mike Innes
e24215ca98 guard test on 1.0 2019-11-15 15:59:42 +00:00
Mike J Innes
665e441919 pkg up 2019-11-15 12:12:28 +00:00
Mike J Innes
9d6f6fdaa3
Merge pull request #926 from janEbert/bc-cuda-fix
Fix binarycrossentropy on CuArrays
2019-11-15 13:05:52 +01:00
Mike J Innes
2471596cdb test on 1.0 2019-11-15 11:50:13 +00:00
dsweber2
89afa20410 Merge branch 'activations' of github.com:dsweber2/Flux.jl into activations 2019-11-14 14:09:27 -08:00
dsweber2
58c794702d simpler test 2019-11-14 14:05:53 -08:00
dsweber2
0fe3ac4e77 bring activations into function call 2019-11-14 13:40:52 -08:00
dsweber2
db92b0e3ce super simple test 2019-11-14 13:40:52 -08:00
dsweber2
6475f6a43e recursive way of doing activations 2019-11-14 13:40:52 -08:00
dsweber2
99679f7e16 deal with empty Chain 2019-11-14 13:40:52 -08:00
dsweber2
d0202a2945 adding the extra commits broke the accumulate version 2019-11-14 13:40:52 -08:00
dsweber2
cdaaca8cfa make activations zygote friendly 2019-11-14 13:40:29 -08:00
Helios De Rosario
ba4e3be0d3
explanations about params in train! 2019-11-14 16:22:31 +01:00
Helios De Rosario
074eb47246
Update training.md 2019-11-12 23:29:38 +01:00
Helios De Rosario
7e1ffd6507
Extend docs about train!
Related to #921: explain why it is not needed to pass the model as argument.
2019-11-08 21:39:00 +01:00
janEbert
a00d8d94ec Add test for CUDA binarycrossentropy 2019-11-08 17:28:54 +01:00