Commit Graph

2400 Commits

Author SHA1 Message Date
bors[bot] d6c53c13db
Merge #1113
1113: Explicitly import `Flux.Optimiser.apply!` in optimiser docs r=dhairyagandhi96 a=SebastianCallh

Closes #1111 

Co-authored-by: Sebastian Callh <sebastian.callh@peltarion.com>
Co-authored-by: Sebastian Callh <sebastian.callh@gmail.com>
2020-04-08 08:14:17 +00:00
Sebastian Callh 7b55bf40a3
Update docs/src/training/optimisers.md
Co-Authored-By: Dhairya Gandhi <dhairya@juliacomputing.com>
2020-04-08 10:11:35 +02:00
Sebastian Callh 58bb72625c Explicitly import `Flux.Optimiser.apply!` in optimiser docs 2020-04-08 07:53:28 +02:00
CarloLucibello 4f693e02cb add model zoo reference 2020-02-29 13:50:23 +01:00
CarloLucibello 4109f2e0d7 cleanup 2020-02-29 13:45:17 +01:00
CarloLucibello 169ed6eb25 add ecosystem 2020-02-29 13:43:03 +01:00
bors[bot] 81a55a0c9e
Merge #1041
1041: add NNlib docs + misc docs improvements r=CarloLucibello a=CarloLucibello

Partially addressing https://github.com/FluxML/NNlib.jl/issues/137.

Also, I'm leaving out the `σ`  activation and using its alias `sigmoid`, since `σ` conveys little information and it is also used to denote a generic activation in the Dense layer. I think we should deprecate `σ` in NNlib, has this been discussed already?

In an ideal world, before merging this, we should get NNlib to either unexport or add docs to its undocumented exports  

Co-authored-by: Carlo Lucibello <carlo.lucibello@gmail.com>
2020-02-29 10:24:39 +00:00
Carlo Lucibello 425fcdbe69 NNlib docs + misc docs improvements 2020-02-29 11:14:48 +01:00
bors[bot] 2dd23574c0
Merge #998
998: test restructure on the GPU r=CarloLucibello a=ChrisRackauckas

Requires https://github.com/FluxML/Zygote.jl/pull/474 to pass

Co-authored-by: Chris Rackauckas <accounts@chrisrackauckas.com>
2020-02-29 09:08:11 +00:00
Dhairya Gandhi 35f6998be7 pkg up 2020-02-27 22:19:06 +05:30
Dhairya Gandhi a121742f9c pkg up 2020-02-27 13:56:05 +05:30
bors[bot] 531d3d4d8b
Merge #1052
1052: update docs and export update! r=dhairyagandhi96 a=CarloLucibello

Fix #951 

Co-authored-by: CarloLucibello <carlo.lucibello@gmail.com>
2020-02-26 19:33:53 +00:00
CarloLucibello 759fe9df2f update docs and export update! 2020-02-26 20:27:39 +01:00
bors[bot] 55616afc11
Merge #960
960: Added utility function outdims to compute output dimensions of a layer r=dhairyagandhi96 a=darsnack

Based on Slack chatter, I added a utility function, `outdims`, that computes the output dimensions for given input dimensions.

Example
```julia
layer = Conv((3, 3), 3 => 16)
outdims(layer, (10, 10)) # returns (8, 8)
```

Co-authored-by: Kyle Daruwalla <daruwalla@wisc.edu>
2020-02-25 17:40:05 +00:00
Dhairya Gandhi 7e58766467
Merge pull request #1047 from MotJuMi/master
Edit description of convolutional layer
2020-02-25 15:39:23 +05:30
Bulat Suleymanov db4eaf254b
Edit description of convolutional layer 2020-02-24 13:16:51 +05:00
Dhairya Gandhi 34ceed5c1f
Merge pull request #1046 from ianshmean/patch-1
Bump Colors compat to include 0.10, 0.11
2020-02-24 10:41:49 +05:30
Ian Butterworth 6ced7e1ecf
expand Colors compat 2020-02-23 13:42:11 -05:00
Kyle Daruwalla f5b9cf659c Updated docs to specify exactly what layers support outdims 2020-02-20 23:38:56 -06:00
Dhairya Gandhi 88b0c65d72
Merge pull request #1035 from matsueushi/remove_get_macro
Remove get! macro
2020-02-20 12:58:16 +05:30
Dhairya Gandhi 8f7a0bb264
Merge pull request #1030 from JuliaTagBot/master
Install TagBot as a GitHub Action
2020-02-19 21:47:31 +05:30
Dhairya Gandhi a38af748e5
Merge pull request #1037 from heliosdrm/heliosdrm-patch-1
update compat to Juno 0.8
2020-02-19 21:46:33 +05:30
bors[bot] e4a84c120f
Merge #1021
1021: nograd for onecold, onehot, onehotbatch r=MikeInnes a=CarloLucibello

fixes #1020 

Co-authored-by: CarloLucibello <carlo.lucibello@gmail.com>
2020-02-17 14:12:48 +00:00
Helios De Rosario 9bb388d953
update Juno compat 2020-02-16 18:29:18 +01:00
Helios De Rosario 6f0710d364
Merge pull request #1 from FluxML/master
update to origin
2020-02-16 18:27:35 +01:00
Viral B. Shah 0b8d1574bf
Merge pull request #984 from aminya/CompatHelper
Adding CompatHelper
2020-02-16 09:44:09 -05:00
matsueushi 6ea7b95384 Remove unused using 2020-02-15 20:06:15 -05:00
bors[bot] fe85a38d78 Merge #1032
1032: Remove outdated reference to truncate! r=dhairyagandhi96 a=mcognetta



Co-authored-by: Marco <mcognetta@users.noreply.github.com>
2020-02-10 08:30:15 +00:00
Marco ae0455517a Remove outdated reference to truncate! 2020-02-10 00:03:11 -08:00
Kyle Daruwalla c37fc3cfa6 Recommitting to trigger build 2020-02-09 19:45:04 -06:00
Julia TagBot d7b20d1a78 Install TagBot as a GitHub Action 2020-02-08 20:02:52 +07:00
CarloLucibello 6499344af3 nograd for onecold, onehot, onehotbatch 2020-02-06 15:41:46 +01:00
bors[bot] 60043fa2aa
Merge #1013
1013: Adapt to GPUArrays/CuArrays changes r=dhairyagandhi96 a=maleadt

Changes in response to a29df67184 and https://github.com/JuliaGPU/CuArrays.jl/pull/576. I suppose the next CuArrays release will need to be breaking because of this.

Maybe the `crossentropy` signature needs to be adjusted to support integer vectors, but I'll leave that decision up to Flux developers. This at least is the quick fix to get the tests passing again.

Co-authored-by: Tim Besard <tim.besard@gmail.com>
2020-02-03 16:29:48 +00:00
Dhairya Gandhi ddc2c20e68
Merge pull request #994 from FluxML/ox/doccustomtraining
Add custom training loops to docs
2020-02-01 11:13:54 +05:30
Tim Besard e2c2ec5575 Don't invoke GPU crossentropy with integers.
Broadcasting log on integers does not work.
2020-01-31 08:22:54 +01:00
Tim Besard e66a7f130f Don't compare CPU with GPU arrays. 2020-01-31 08:22:21 +01:00
Dhairya Gandhi 620cffc45c
Merge pull request #1008 from FluxML/tb/cuindex
Remove unused imports.
2020-01-29 18:52:53 +05:30
Tim Besard d88f63adb4 Remove unused imports. 2020-01-29 12:15:41 +01:00
Chris Rackauckas 9803826a36 test restructure on the GPU
Requires https://github.com/FluxML/Zygote.jl/pull/474
2020-01-20 13:53:28 -05:00
Lyndon White 7797e31b44
Add custom training loops to docs 2020-01-16 21:57:59 +00:00
bors[bot] d1edd9b16d
Merge #680
680: Added new loss functions. r=thebhatman a=thebhatman

I have added the KL Divergence Loss function, Poisson loss function, Logcosh loss, and Hinge loss function.

Co-authored-by: Manjunath Bhat <manjunathbhat9920@gmail.com>
Co-authored-by: thebhatman <manjunathbhat9920@gmail.com>
2020-01-13 15:46:25 +00:00
Manjunath Bhat 747e01ea02
Test to check for spurious promotions 2020-01-13 18:33:30 +05:30
Dhairya Gandhi 048c31f609 bump Flux version to v0.10.1 2020-01-13 18:16:29 +05:30
bors[bot] f7f0ebbffd
Merge #992
992: Compat bounds for a couple more packages r=dhairyagandhi96 a=dhairyagandhi96

adds compatibility bounds for a few more packages

cc @MikeInnes 

Co-authored-by: Dhairya Gandhi <dhairya@juliacopmuting.com>
2020-01-13 12:29:05 +00:00
Dhairya Gandhi cd4626d5a7 compat bounds for a couple more packages 2020-01-13 17:38:59 +05:30
bors[bot] 2b222b15fa
Merge #991
991: Update CuArrays + Zygote deps  r=dhairyagandhi96 a=dhairyagandhi96

cc @MikeInnes 

Co-authored-by: Dhairya Gandhi <dhairya@juliacopmuting.com>
2020-01-13 11:14:21 +00:00
Dhairya Gandhi e1698e6617 up cuarrays 2020-01-13 16:18:20 +05:30
Dhairya Gandhi e2a97aec24 up cuda+zygote deps 2020-01-13 16:16:24 +05:30
Dhairya Gandhi de40476beb doc tests on julia 1.3 2020-01-13 14:10:34 +05:30
Dhairya Gandhi d7953ff573 test on julia 1.3+ 2020-01-13 13:45:40 +05:30