Skip to content

GNNGraphs-v1.3.0
Compare
Choose a tag to compare
@github-actions github-actions released this 30 Nov 18:52
GNNGraphs-v1.3.0
e350613
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

GNNGraphs GNNGraphs-v1.3.0

Merged pull requests:

Closed issues:

  • Differences to GeometricFlux.jl? (#2)
  • add benchmarks (#4)
  • register package (#5)
  • add documentation (#6)
  • add examples (#7)
  • implement graph concatenation (#12)
  • improve documentation (#14)
  • pretty print GNNGraph (#23)
  • define a message_and_aggregate method (#29)
  • index not displayed in documentation pages (#32)
  • A Logo is needed (#35)
  • add support to edge weight in GCNConv (#40)
  • TagBot trigger issue (#44)
  • Problem with GNNChain and NNConv (#49)
  • graph NeuralODE example not working on gpu (#56)
  • Adding a GATv2 layer (#74)
  • Move the package to the FluxML org (#80)
  • Explainer vs GeometricFlux (#81)
  • Weights not included in GNNGraph made from SimpleWeightedDiGraph (#85)
  • Hash function for GNNGraph (#87)
  • no method matching getobs(::NamedTuple{(:x,), Tuple{Matrix{Float32}}} (#88)
  • Custom Function GPU Compatibitlity Issue: Indexing (#91)
  • Flux.batch Overloading for Generators (#92)
  • outputsize for GNNChain (#96)
  • Problem with InlineStrings.jl (#98)
  • batching scales quadratically (#99)
  • Include undirected graphs (#101)
  • Merging multiple feature arrays (#102)
  • Implement add_reverse_edges (#103)
  • conflict with CSV and GNNGraphs when running Flux.batch (#104)
  • propagate() is 20x slower than built-in sparse matmul (#106)
  • add fusing propagate specialization for e_mul_xj (#108)
  • implement set_edge_weights (#110)
  • Question about temporal graph neural networks (#112)
  • Gradient of edge weights is nothing with fused e_mul_xj (#113)
  • aggregate_neighbors() is 100x slower than equivalent sparse matrix operation (#124)
  • Failure to combine SparseDiffTools.autoback_hesvec and GCNConv (#125)
  • GATv2Conv show method errors (#126)
  • Support edge attributes in as many layers as possible (#128)
  • Roadmap to merge GeometricFlux.jl and GraphNeuralNetworks.jl (#132)
  • Failed to compile PTX code (#133)
  • GINConv not working on GPU when not all nodes are connected (#138)
  • Batchnorm for Integers after GCNConv or GINConv on GPU (#140)
  • Slow interaction with DataLoader (#141)
  • Heterogeneous graphs support (#144)
  • hope can load dataset with GraphMLDatasets.jl (#149)
  • GPU memory filling up (#150)
  • Citing GraphNeuralNetworks.jl (#151)
  • GSoC 2022 (#157)
  • Open Graph Benchmark integration (#162)
  • message passing for multiple feature arrays (#166)
  • Construct Graphs.SimpleDiGraph graphs from GNNGraphs (#167)
  • Add tutorials written in Pluto (#168)
  • Info about features in the compact show (#169)
  • don't automatically batch when getobs from an array of graphs (#170)
  • Inaccurate GATv2Conv Documentation (#175)
  • add radius_graph api (#177)
  • add show methods for WithGraph and GNNChain (#178)
  • knn_graph yields same results with or without self loops (#179)
  • Missing bounds checking when working on GPU (#181)
  • Spelling Error of edge? (#184)
  • Duplicate indexes in documents (#187)
  • Implement GraphWorld for fake graphs benchmarking (#188)
  • Implement Learnable Structural Positional Encoding (LSPE) (#190)
  • copy(::GNNGraph)? (#191)
  • Some design issues (#193)
  • Support for Heterogeneous Graphs (#199)
  • Flux destructure/restructure bug (#200)
  • Remove hard dependency on GraphNeuralNetworks from Pluto Notebooks (#204)
  • NNConv tests are failing (#208)
  • Allow for additonal features in GNNGraph (#210)
  • Question about the GCNConv layer code (#211)
  • Support for multiple graphs in GNNGraph (#212)
  • Formatting errors in the tutorial (#213)
  • Generation of documentation is very slow because of Pluto (#227)
  • GNN.jl in tutorials (#228)
  • Julia Formatter (#238)
  • GNNGraph with multiple edge features not working (#243)
  • Overflows in GATConv and GATv2Conv (#246)
  • Convolutions for GNNHeteroGraphs (#254)
  • Update documentation: Convolutional Layers (#255)
  • Dropout inside GATConv layer (#258)
  • Bad performance of GCNConv (#259)
  • Documentation link (#262)
  • GCNConv without normalization (#277)
  • add_edges adds a non-existent edge to its DataStore (#280)
  • Graph classification: multiple graphs associated with a common label (#282)
  • need more informative error for dimension mismatch (#283)
  • convenience feature setter (#284)
  • @functor default for GNNLayers (#288)
  • add docs for HeteroGraphConv (#302)
  • add batching for GNNHeteroGraph (#303)
  • Local pooling in graph regression/classification problems (#307)
  • use extension instead of CUDA hard dependence (#317)
  • Update Flux.trainable (#323)
  • AGNNConv behaviour different from mathematical definition (due to self loops) (#325)
  • implement add_self_loops(g, edge_t) for heterographs (#329)
  • Edge weights not properly documented for GNNHeteroGraphs (and implement new function to add new edge weights?) (#331)
  • HeteroGraphConv bug: ERROR: duplicate field name in NamedTuple: "movie" is not unique (#332)
  • add_edges for GNNHeteroGraph does not allow providing the number of nodes (#334)
  • add empty constructor for heterograph (#338)
  • The constraint in Flux.batch(gs::AbstractVector{<:GNNHeteroGraph}) does not seem to be strong enough (#341)
  • Error in Example Code (#346)
  • Outdated default package installation (#348)
  • View arrays on GPU cause scalar indexing error (#349)
  • documentation proposal (#357)
  • support Lux (#372)
  • getgraph not working on GPU (#377)
  • Example given for GNNGraph results in error (#380)
  • Error in one of the examples in the Working with GNNGraph page (#401)
  • Duplicated method definitions of GINConv (#406)
  • turn this into a monorepo (#433)
  • Maybe state difference with GeometricFlux.jl. (#435)
  • plan for splitting the package (#450)
  • use Flux.@layer instead of Flux.@functor (#452)
  • consider using MultiDocumenter (#456)
  • reinstate temporal graphs tutorials (#457)
  • random graph generators should take an rng instead of a seed (#459)
  • plan for GNNLux.jl (#461)
  • Cannot create GNNGraph with unconnected nodes (#472)
  • Implementation of recommender system based on GNN (#473)
  • GNNs.jl's CI is failing for GRAPH_T = :dense (#476)
  • move all tests to TestItems.jl and TestItemsRunner.jl (#477)
  • document the monorepo structure and the package dependencies (#483)
  • GCNConv layer fails when the GNNGraph comes from an adjacency matrix (#486)
  • move GraphNeuralNetworks.jl to its own folder (#495)
  • Comparison to GeometricFlux.jl (#502)
  • Overriding Base.getproperty(vds::Vector{DataStore}, s::Symbol) conflicts A.ref usage in julia (#504)
  • move repo to JuliaGraphs org (#506)
  • move NeighborLoader to GNNGraphs (#507)

Contributors

functor, CarloLucibello, and 5 other contributors
Assets 2
Loading