FluxML/Flux.jl
Ver no GitHubWeights shape not validated against kernel, channels
Open
#2.506 aberto em 25 de out. de 2024
good first issuehelp wanted
Métricas do repositório
- Stars
- (4.725 stars)
- Métricas de merge de PR
- (Mesclagem média 4h 27m) (2 fundiu PRs em 30d)
Description
weights = Flux.kaiming_normal()(3, 3, 1)
Conv((3, 3), 1 => 1; pad = (1, 1), init = (_...) -> weights)
# Conv((3,), 3 => 1, pad=1) # 10 parameters
weights = Flux.kaiming_normal()(3, 3, 1, 1)
Conv((3, 3), 1 => 1; pad = (1, 1), init = (_...) -> weights)
# Conv((3, 3), 1 => 1, pad=1) # 10 parameters
I wanted to strictly specify the weight init for testing, but encountered this odd result. I think there should be validation to ensure that the weight shape matches the kernel size and input channels, and error if there is a mismatch.