Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / FluxML/Flux.jl issues and pull requests

#2172 - CompatHelper: bump compat for CUDA to 4, (keep existing compat)

Pull Request - State: closed - Opened by github-actions[bot] almost 2 years ago - 10 comments

#2171 - On some recent Flux experiences

Issue - State: open - Opened by mcabbott almost 2 years ago - 4 comments
Labels: documentation

#2170 - Update GH Actions across all repos

Issue - State: closed - Opened by ToucheSir almost 2 years ago - 9 comments
Labels: good first issue

#2165 - SSIM loss

Issue - State: closed - Opened by nikopj almost 2 years ago - 10 comments
Labels: enhancement

#2164 - Feature Request: Load pre-trained weights (.pth files etc.) into a flux model

Issue - State: open - Opened by ejmeitz almost 2 years ago - 6 comments
Labels: enhancement

#2163 - 0.2 tutorial at the top of google for "flux mnist"

Issue - State: closed - Opened by anandijain almost 2 years ago - 4 comments
Labels: help wanted, documentation

#2156 - Match layer output to weights

Pull Request - State: closed - Opened by mcabbott almost 2 years ago - 6 comments
Labels: run downstream test

#2155 - Hoping to offer a version without cuda

Issue - State: closed - Opened by NeutronT almost 2 years ago - 13 comments
Labels: cuda

#2151 - Use `NNlib.bias_act!`

Pull Request - State: closed - Opened by mcabbott almost 2 years ago

#2147 - CompatHelper: bump compat for MLUtils to 0.4, (keep existing compat)

Pull Request - State: closed - Opened by github-actions[bot] almost 2 years ago - 1 comment

#2146 - MultiHeadAttention implementation

Pull Request - State: closed - Opened by CarloLucibello almost 2 years ago - 16 comments

#2143 - remove some warnings in the test outputs

Pull Request - State: closed - Opened by CarloLucibello almost 2 years ago - 6 comments

#2139 - Remove greek-letter keyword arguments

Pull Request - State: closed - Opened by mcabbott almost 2 years ago - 3 comments
Labels: run downstream test

#2133 - Delete bors.toml

Pull Request - State: closed - Opened by CarloLucibello almost 2 years ago - 4 comments

#2132 - [WIP] Move CUDA support to a package extension

Pull Request - State: closed - Opened by IanButterworth almost 2 years ago - 32 comments
Labels: breaking

#2123 - fix train! test

Pull Request - State: closed - Opened by CarloLucibello almost 2 years ago - 1 comment

#2117 - fix cpu(x) for immutable arrays

Pull Request - State: closed - Opened by CarloLucibello almost 2 years ago - 1 comment

#2092 - Add SnoopPrecompile

Pull Request - State: closed - Opened by mcabbott about 2 years ago - 2 comments

#2077 - `rrule` for `Recur`

Pull Request - State: closed - Opened by mcabbott about 2 years ago
Labels: gradients, RNN

#2074 - Add GH action format check files with `JuliaFormatter`

Pull Request - State: closed - Opened by Saransh-cpp about 2 years ago - 15 comments

#2065 - Make `rng_from_array` non differentiable

Pull Request - State: closed - Opened by Saransh-cpp about 2 years ago - 3 comments

#2059 - Fix the failing CI build

Pull Request - State: closed - Opened by christiangnrd about 2 years ago - 3 comments

#2044 - Macro to display model struct the way Flux does

Issue - State: closed - Opened by ericphanson about 2 years ago - 2 comments

#2040 - Pull request #2007 causes Flux.params() calls to not get cached

Issue - State: open - Opened by christiangnrd over 2 years ago - 10 comments

#2031 - Add `EmbeddingBag`

Pull Request - State: closed - Opened by mcognetta over 2 years ago - 11 comments
Labels: enhancement

#2028 - RFC: add a supertype to layers

Pull Request - State: closed - Opened by mcabbott over 2 years ago - 5 comments
Labels: discussion

#2021 - Add a logistic regression example to the Getting Started section

Pull Request - State: closed - Opened by Saransh-cpp over 2 years ago - 2 comments
Labels: documentation

#2013 - Functional equivalents missing for some layers

Issue - State: open - Opened by theabhirath over 2 years ago - 3 comments
Labels: enhancement, good first issue

#1986 - Deprecate Flux.Optimisers and implicit parameters in favour of Optimisers.jl and explicit parameters

Issue - State: open - Opened by CarloLucibello over 2 years ago - 17 comments
Labels: discussion, optimisers-dot-jl, deprecation

#1982 - No error from negative learning rates

Issue - State: closed - Opened by KronosTheLate over 2 years ago - 20 comments
Labels: help wanted, optimisers-dot-jl

#1969 - CompatHelper: bump compat for ArrayInterface to 6, (keep existing compat)

Pull Request - State: closed - Opened by github-actions[bot] over 2 years ago - 1 comment

#1966 - PINN loss doesn't converge to 0?

Issue - State: closed - Opened by a-pulido over 2 years ago - 9 comments
Labels: gradients

#1961 - Very slow "using Flux"

Issue - State: closed - Opened by mattiadg over 2 years ago - 21 comments
Labels: help wanted, performance

#1932 - Add a macro to opt-in to fancy printing, and to everything else

Pull Request - State: closed - Opened by mcabbott over 2 years ago - 12 comments
Labels: discussion, run downstream test

#1907 - Choosing model serialization format(s) for cross-framework support (like HuggingFace)

Issue - State: closed - Opened by darsnack over 2 years ago - 9 comments
Labels: discussion

#1855 - Block parameters for RNNs (for cudnn paths)

Pull Request - State: closed - Opened by mkschleg almost 3 years ago - 11 comments
Labels: RNN

#1829 - Roadmap

Issue - State: open - Opened by darsnack almost 3 years ago - 11 comments

#1793 - Allow NamedTuples in `batchmemaybe`

Pull Request - State: closed - Opened by DhairyaLGandhi almost 3 years ago - 15 comments

#1790 - Added the wrapper `Bidirectional` for RNN layers

Pull Request - State: open - Opened by dcecchini almost 3 years ago - 13 comments
Labels: enhancement, RNN

#1789 - Update basics.md to include gradient docs

Pull Request - State: closed - Opened by logankilpatrick almost 3 years ago - 10 comments
Labels: documentation

#1755 - Remove Recurrent cell eltype restriction

Pull Request - State: closed - Opened by rmsrosa about 3 years ago - 13 comments
Labels: RNN

#1682 - add NamedTuple chain constructor

Pull Request - State: closed - Opened by DhairyaLGandhi over 3 years ago - 5 comments

#1664 - Allow user to change the channel axis for `BatchNorm` function and the likes

Issue - State: open - Opened by vboussange over 3 years ago - 22 comments

#1606 - Allow BatchNorm training on CUDA with `track_stats=false`

Issue - State: open - Opened by ToucheSir over 3 years ago - 4 comments
Labels: help wanted, good first issue

#1592 - Tied weights using Flux layers

Issue - State: open - Opened by dfenn over 3 years ago - 9 comments

#1566 - Support multiple GPU backends

Pull Request - State: closed - Opened by jpsamaroo over 3 years ago - 26 comments

#1509 - Fix Norm Layers, Again

Pull Request - State: closed - Opened by DhairyaLGandhi almost 4 years ago - 30 comments

#1474 - CompatHelper: bump compat for "Functors" to "0.2"

Pull Request - State: closed - Opened by github-actions[bot] almost 4 years ago - 12 comments

#1470 - remove dataset tests

Pull Request - State: closed - Opened by CarloLucibello almost 4 years ago - 3 comments

#1452 - CompatHelper: bump compat for "Adapt" to "3.0"

Pull Request - State: closed - Opened by github-actions[bot] almost 4 years ago - 3 comments

#1451 - release 0.11.4

Pull Request - State: closed - Opened by CarloLucibello almost 4 years ago - 7 comments

#1438 - CompatHelper: bump compat for "Reexport" to "1.0"

Pull Request - State: closed - Opened by github-actions[bot] almost 4 years ago - 6 comments

#1435 - CompatHelper: bump compat for "Zygote" to "0.6"

Pull Request - State: closed - Opened by github-actions[bot] almost 4 years ago

#1431 - PyTorch feature parity

Issue - State: open - Opened by CarloLucibello almost 4 years ago - 94 comments
Labels: help wanted

#1413 - Add buildkite pipeline

Pull Request - State: closed - Opened by DhairyaLGandhi almost 4 years ago - 3 comments

#1393 - remove implicit conversions

Pull Request - State: closed - Opened by CarloLucibello about 4 years ago - 40 comments

#1385 - TagBot trigger issue

Issue - State: closed - Opened by JuliaTagBot about 4 years ago - 57 comments

#1378 - CompatHelper: bump compat for "CUDA" to "2.1"

Pull Request - State: closed - Opened by github-actions[bot] about 4 years ago

#1376 - CompatHelper: bump compat for "CUDA" to "2.1"

Pull Request - State: closed - Opened by github-actions[bot] about 4 years ago

#1365 - RNN design for efficient CUDNN usage

Issue - State: open - Opened by jeremiedb about 4 years ago - 9 comments

#1344 - CompatHelper: bump compat for "CUDA" to "2.0"

Pull Request - State: closed - Opened by github-actions[bot] about 4 years ago - 3 comments

#1304 - Metal GPU acceleration on Apple Silicon

Issue - State: closed - Opened by BooBSD over 4 years ago - 5 comments

#1067 - [WIP] Added some Callback functions

Pull Request - State: closed - Opened by AdarshKumar712 over 4 years ago - 9 comments

#1038 - CompatHelper: bump compat for "Colors" to "0.11"

Pull Request - State: closed - Opened by github-actions[bot] almost 5 years ago - 2 comments

#969 - add lookahead optimiser

Pull Request - State: closed - Opened by chengchingwen almost 5 years ago - 9 comments

#943 - Fixes #900

Pull Request - State: closed - Opened by DhairyaLGandhi almost 5 years ago - 5 comments

#941 - Don't include the CUDA module during precompilation.

Pull Request - State: closed - Opened by maleadt almost 5 years ago - 3 comments

#932 - Travis: test on 1.0

Pull Request - State: closed - Opened by MikeInnes about 5 years ago - 4 comments

#924 - CUDA package initialization improvements

Pull Request - State: closed - Opened by maleadt about 5 years ago - 2 comments

#922 - Restore Julia 1.0 compatibility.

Pull Request - State: closed - Opened by maleadt about 5 years ago - 2 comments
Labels: cuda

#882 - Check if CUDA availability changed during init.

Pull Request - State: closed - Opened by maleadt about 5 years ago - 4 comments
Labels: cuda

#874 - Move CUDNN wrappers to CuArrays

Pull Request - State: closed - Opened by MikeInnes about 5 years ago - 2 comments
Labels: cuda

#780 - How to have a stable GPU memory while being performant?

Issue - State: closed - Opened by ByzanTine over 5 years ago - 6 comments

#707 - accumulate gradient with the new gradient API?

Issue - State: closed - Opened by chengchingwen over 5 years ago - 10 comments
Labels: enhancement

#663 - Add PLSTM and FCLSTM layer

Pull Request - State: closed - Opened by swiesend over 5 years ago - 9 comments

#543 - Mixed precision training.

Issue - State: closed - Opened by KristofferC almost 6 years ago - 5 comments
Labels: float16

#465 - Destroy AD graph when doing in-place gradients

Pull Request - State: closed - Opened by MikeInnes about 6 years ago - 3 comments

#395 - Add max/mean unpooling operations

Issue - State: open - Opened by AzamatB about 6 years ago - 5 comments
Labels: enhancement, help wanted, good first issue

#173 - OpenCL support

Issue - State: closed - Opened by MikeInnes almost 7 years ago - 10 comments
Labels: help wanted

#115 - Fixed the spectral normalization

Pull Request - State: closed - Opened by pevnak almost 7 years ago - 5 comments