Dhairya Gandhi
df22bc5c8f
removed argument from stop function
2018-08-20 14:02:09 +05:30
Dhairya Gandhi
06db6ed314
housekeeping: fixing typo
2018-08-20 13:48:28 +05:30
Dhairya Gandhi
394b4167ce
moving stop to Optimise
2018-08-20 13:43:08 +05:30
Dhairya Gandhi
06aad375fc
properly importing functions
2018-08-20 13:35:55 +05:30
Dhairya Gandhi
e239eb1105
properly importing functions
2018-08-20 13:30:05 +05:30
Dhairya Gandhi
1228e9c5e2
removed include statement
2018-08-19 22:55:14 +05:30
Dhairya Gandhi
9c98272cf0
catching exception
2018-08-19 17:38:00 +05:30
Dhairya Gandhi
257e2a7d2e
checking exception
2018-08-19 17:11:11 +05:30
Dhairya Gandhi
5c42c8689c
printing expception
2018-08-19 17:04:31 +05:30
Dhairya Gandhi
fbd82a6925
added end
2018-08-19 15:19:45 +05:30
Dhairya Gandhi
8229c8e045
modified training loop
2018-08-19 15:17:07 +05:30
Dhairya Gandhi
2aa057ec08
fixed throwing exception
2018-08-19 14:54:54 +05:30
Dhairya Gandhi
887bfad312
returning :stop
2018-08-18 08:28:47 +05:30
Dhairya Gandhi
65a5ecccd2
returning
2018-08-18 08:24:49 +05:30
Dhairya Gandhi
999b00b64d
fixed typo
2018-08-17 19:45:10 +05:30
Dhairya Gandhi
0524964400
fixed typo
2018-08-17 19:40:48 +05:30
Dhairya Gandhi
24a3bce495
added stop to break training loop
2018-08-17 17:46:13 +05:30
Josh Christie
c8307a0627
Use @info for logging
2018-08-11 14:42:33 +01:00
Avik Pal
5db7a3a3ad
Fix Optimizers
2018-08-11 18:23:47 +05:30
pevnak
3510c837a8
zeros replaced by zero
2018-08-03 15:14:25 +01:00
Jarvist Moore Frost
344a750770
Merge branch 'master' of github.com:jarvist/Flux.jl into HEAD
2018-07-03 11:15:43 +01:00
Jarvist Moore Frost
aee4a83c55
Add ADAMW weight-decay.
...
See http://www.fast.ai/2018/07/02/adam-weight-decay/ and the original
paper https://arxiv.org/abs/1711.05101.pdf for context.
I don't know what I'm doing, and this is quite possibly wrong - but on
a simple Char-RNN I have lying around on my harddisk, this seems to
improve the rate of learning consistently for different hyperparameters
vs. standard ADAM with the same decay constant.
2018-07-03 11:11:32 +01:00
Mike J Innes
2723c9ee04
Merge pull request #257 from staticfloat/sf/back_inf_nan
...
Check for `Inf` and `NaN` within `back!(::TrackedReal)`
2018-06-26 14:42:33 +01:00
Tejan Karmali
4a24b69976
Merge branch 'master' into nadam-opt
2018-06-08 16:54:41 +05:30
staticfloat@gmail.com
9fdbe843ef
Check for Inf
and NaN
within back!(::TrackedReal)
...
This is often checked for within user code, no reason to do that, let's
do it for them within `back!(::TrackedReal)`
2018-05-07 15:30:44 -07:00
CarloLucibello
e186b958dd
more exports
2018-05-01 12:13:14 +01:00
Sujeet Akula
8c042bd522
element wise max()
2018-04-26 21:12:31 +10:00
Sujeet Akula
5e5f255f81
export typo
2018-04-26 17:42:04 +10:00
Sujeet Akula
4586bda5ab
export/test adamax
2018-04-26 17:40:11 +10:00
Sujeet Akula
b6508e2416
add adamax
2018-04-26 17:37:24 +10:00
tejank10
65847bb745
moved epsilon into sqrt
2018-04-04 15:25:20 +05:30
tejank10
3ead662987
Update rule fixed
2018-04-04 15:18:44 +05:30
tejank10
ea9b5471fa
NADAM optimizer
2018-04-03 01:27:22 +05:30
Mike Innes
bfd6a4c0ec
cleaner interrupts
2018-03-05 23:05:45 +00:00
Mike Innes
5153cde847
move epochs
2018-03-05 22:56:22 +00:00
Mike J Innes
79e4e25fea
seperate number type
2018-02-07 20:39:36 +00:00
Mike J Innes
282889970d
seperate tracking infrastructure from array wrapper
2018-02-07 17:43:25 +00:00
Mike J Innes
5b97d2ba04
closes #127
2017-12-13 18:24:56 +00:00
Mike J Innes
24a6569589
Merge branch 'master' into amsgrad
2017-12-08 18:20:53 +00:00
baggepinnen
41febee9c1
Export and indent
2017-12-04 09:34:27 +01:00
baggepinnen
36001d085a
Implement AMSGrad optimiser
2017-12-04 09:17:05 +01:00
CarloLucibello
13b934c250
improve optimizers
2017-11-24 12:12:20 +01:00
Mike J Innes
9f5c4dd3e9
Merge pull request #104 from baggepinnen/patch-1
...
Allow array of optimisers to train!
2017-11-21 17:16:35 +01:00
Mike J Innes
979949d01a
style
2017-11-21 15:25:09 +01:00
Fredrik Bagge Carlson
8991ce028c
Fix bug in rmsprop and adadelta
...
`@. p.Δ = η * p.Δ / √acc` parses correctly while `@. p.Δ /= √acc*η` seems to parse like `@. p.Δ /= (√acc*η)`, hence the step size was de facto interpreted as `1/η`
2017-11-14 17:32:16 +01:00
Fredrik Bagge Carlson
97244e0a68
Allow array of optimisers to train!
...
This allows an array of optimisers to be sent to `train!`
2017-11-04 13:27:32 +01:00
Mike J Innes
99a7697d13
adam eta default arg
2017-10-19 14:31:34 +01:00
Mike J Innes
5b6a5667ed
tracked array restructure
2017-10-18 22:54:58 +01:00
Mike J Innes
07ad7cfa40
learning rate as default arg
2017-10-18 17:07:49 +01:00
Mike J Innes
7426faf37d
optimiser docs
2017-10-18 12:09:48 +01:00
CarloLucibello
041079237e
add docsting to train!
2017-10-17 21:04:18 +01:00
CarloLucibello
6d3a2a2210
change argument name for better clarity
2017-10-17 21:04:18 +01:00
GenaBitu
ef6d10886d
Exposed all optimisers
2017-10-06 14:20:09 +01:00
pevnak
bfcc1ac25d
exposing optimisers
2017-10-05 12:36:18 +01:00
Mike J Innes
5fd1b7d9a2
remove gc hack
2017-10-02 20:50:18 +01:00
Mike J Innes
7c8dba0b85
gc in training loop
2017-09-27 23:14:58 +01:00
Mike J Innes
120a6db2bb
Merge branch 'master' of github.com:MikeInnes/Flux.jl
2017-09-27 21:16:23 +01:00
Mike J Innes
4bafa2b374
generic tree functions
2017-09-27 21:11:21 +01:00
Mike J Innes
94e38c05b8
more informative
2017-09-27 18:33:23 +01:00
Mike J Innes
f2052739c1
tweaks
2017-09-12 14:11:03 +01:00
Mike J Innes
3f83be7bb7
more flexible training loop
2017-09-11 13:11:55 +01:00
Mike J Innes
085d3aa9b4
handle epoch elsewhere
2017-09-07 00:29:55 -04:00
Mike J Innes
aeaa138b6d
cb convenience
2017-09-07 00:27:16 -04:00
Mike J Innes
cca4d25a10
efficient traversal
2017-09-06 23:09:32 -04:00
Mike J Innes
47ba702747
tweak optimiser interface
2017-09-03 17:10:04 -04:00
Mike J Innes
e57ae77bbb
juno progress
2017-09-03 02:44:32 -04:00
Mike J Innes
fe2b35facc
add callbacks back
2017-09-01 23:59:44 -04:00
Mike J Innes
bf098d551c
fuck
2017-09-01 23:41:44 -04:00
Mike J Innes
387686eb41
optimisers rework
2017-09-01 17:06:51 -04:00
Mike J Innes
b95dae1868
opt refactor
2017-08-31 14:55:23 -04:00
ylxdzsw
97ecb26003
wip optimisers
2017-08-29 17:00:24 -04:00
Mike J Innes
12dc6b66c5
whoops
2017-08-24 22:23:05 +01:00
Mike J Innes
e7f26370d7
training tweaks
2017-08-24 16:10:04 +01:00
Mike J Innes
1526b13691
basic training loop
2017-08-24 11:42:29 +01:00
Mike J Innes
bafecfede1
sgd
2017-08-22 22:25:18 +01:00
Mike J Innes
0ce8c0cee4
param collection
2017-08-22 17:13:03 +01:00