GNNLux GATConv/GATv2Conv initialize the attention vector `a` without the setup rng, so it's identical for every seed

未关闭 适合新手
  • 哪里不对？
#736 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 2 天内回复

还没有人认领这个 Issue。

评估

难度
2/5
预计耗时
1-3 小时
新手友好度
76/100
Issue 类型
缺陷
描述清晰度
描述清楚
活跃度
活跃
技术栈
julia
领域
machine-learning

调研方向

Issue 指向 GNNLux/src/layers/conv.jl 中 GATConv 和 GATv2Conv 的初始化器，大约在第 901 行和第 1058 行。阅读这些构造函数，并比较 GNNLux 的其他层如何将 setup RNG 传给 init_weight；issue 的最小示例展示了如何检查不同 seed 下的行为，以及如何使用仅接收 RNG 的初始化器进行检查。完成标准是两个 attention 向量都使用提供的 RNG，并且示例不再报告不同 seed 下的向量相同。

由索引模型根据 Issue 内容生成。

描述

In GNNLux, GATConv and GATv2Conv create the attention parameter a without passing the rng given to initialparameters:

https://github.com/JuliaGraphs/GraphNeuralNetworks.jl/blob/54025c6257a1e00a8521a6197dd793ce3941a8cc/GNNLux/src/layers/conv.jl#L901

https://github.com/JuliaGraphs/GraphNeuralNetworks.jl/blob/54025c6257a1e00a8521a6197dd793ce3941a8cc/GNNLux/src/layers/conv.jl#L1058

a = l.init_weight(ein > 0 ? 3out : 2out, l.heads)   # GATConv
a = l.init_weight(out, l.heads)                     # GATv2Conv

With the default glorot_uniform, the rng-less WeightInitializers method falls back to its internal fixed Xoshiro(1234), so a comes out the same no matter which rng is passed to Lux.setup. Every GAT layer with the same shape therefore starts with an identical attention vector, and changing the seed (e.g. for multiple runs or an ensemble) doesn't change it. A custom init_weight that only has an (rng, dims...) method would also fail here with a MethodError.

Minimal example
using GNNLux, Lux, Random

for L in (GATConv(3 => 4; heads = 2), GATv2Conv(3 => 4; heads = 2))
    p0 = Lux.setup(Xoshiro(0), L)[1]
    p1 = Lux.setup(Xoshiro(1), L)[1]
    println(nameof(typeof(L)), ": a equal across seeds = ", p0.a == p1.a,
            ", first Dense weight equal across seeds = ", first(values(p0)).weight == first(values(p1)).weight)
end

# init_weight is called without an rng for `a`
calls = Any[]
myinit(rng::AbstractRNG, dims::Integer...) = (push!(calls, (typeof(rng), dims)); zeros(Float32, dims...))
myinit(dims::Integer...) = (push!(calls, ("no rng", dims)); zeros(Float32, dims...))
Lux.setup(Xoshiro(0), GATv2Conv(3 => 4; heads = 2, init_weight = myinit))
println(calls)

Output (Julia 1.13.3, GNNLux 0.3.0, Lux 1.31.6, WeightInitializers 1.3.5, CPU):

GATConv: a equal across seeds = true, first Dense weight equal across seeds = false
GATv2Conv: a equal across seeds = true, first Dense weight equal across seeds = false
Any[(Xoshiro, (8, 3)), (Xoshiro, (8, 3)), ("no rng", (4, 2))]

The Dense sublayers react to the seed, but a doesn't. The two lines above are unchanged on current master (54025c6), so I expect the same behaviour there, though I ran the example against the registered 0.3.0.

Expected

a should be drawn from the rng passed to initialparameters, i.e. l.init_weight(rng, ...), like the other GNNLux layers (GCNConv, GMMConv, NNConv, ...).

Related (smaller)

In GNNLux GATConv, dense_x and dense_e are built as Dense(... , use_bias = false) without passing init_weight (L888-889), so a custom init_weight only affects a. GATv2Conv passes init_weight to its dense_i/dense_j/dense_e. Happy to split this into a separate issue if you prefer.

主要语言
Julia
星标
312
派生
76
平均合并
2 天 19 小时
30 天内合并 PR
13

环境准备

这个项目没有提供开发容器、Dockerfile 或贡献指南，环境需要你自己搭建：先看它的 README，通用步骤见我们的新手贡献指南。

从这里开始

  1. 先读完整个 Issue，再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库，在一个分支上完成修改。
  4. 提交 Pull Request，并在描述里引用这个 Issue 编号。

JuliaGraphs/GraphNeuralNetworks.jl 的其他 Issue

查看 JuliaGraphs/GraphNeuralNetworks.jl 的全部 Issue

相似的 Issue

更多 Julia Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。