andevellicus
andevellicus
I'm also running into issues where when I change the structure of the package, JLD2 gets extremely fussy -- which when I'm playing with different models and changing some of...
Could 3D functions (5D tensors) please be supported?
At the moment KnetLayers' implementation of CNNs is restricted to 2D inputs: ``` function Filtering{T}(;height::Integer, width::Integer, inout::Pair=1=>1, activation::ActOrNothing=NonAct(), winit=xavier, binit=zeros, atype=arrtype, opts...) where T
Not sure if this would be helpful, but I implemented this in my code: ``` # callbacks.jl using DelimitedFiles # Struct to hold the training state struct TrainingState current_epoch::Int model::Model...
Similar issue. I have TF 2.1, ran keras2onnx on my model, and then tried to load into Julia. Got conv3d_0/kernal:0 not found.
Same here, would love to use this but currently won't run....
I'd be happy to give it a go, just not sure how to go about doing that. I'm using the Julia bindings, if that makes a difference. Also, I think...
I'm new to Julia, so correct me if I'm wrong, but maybe using multiple dispatch based on the kernel size passed? I used your code to implement it in my...
Any luck with this?