KernelFunctions.jl icon indicating copy to clipboard operation
KernelFunctions.jl copied to clipboard

Julia package for kernel functions for machine learning

Results 126 KernelFunctions.jl issues
Sort by recently updated
recently updated
newest added

I looked through existing issues and didn't find someone suggesting this before. But you could in principle replace the special case matern functions by a parametrized type. The advantage is...

**Summary** This PR introduces functionality for lazily representing kernel matrices, which is necessary when the matrix might be too large to store in memory. Fixes #514 **Proposed changes** * new...

Adressess https://github.com/JuliaGaussianProcesses/KernelFunctions.jl/issues/506. Implements a new kernel composition `KernelTensorSum` and related `⊕` operator. The naming should be discussed since the meaning of `KernelTensorSum` might not be appropriate. Suggestions are welcome.

In some cases we do not want to build the kernel matrix explicitely but simply want to evaluate its components when necessary (for example in `K * v`). We should...

enhancement

**Summary** Enable general output indices, e.g. symbols: ```julia julia> KernelFunctions.MOInputIsotopicByFeatures([1,2,3], [:a, :b]) 6-element KernelFunctions.MOInputIsotopicByFeatures{Int64, Vector{Int64}, Symbol, Vector{Symbol}}: (1, :a) (1, :b) (2, :a) (2, :b) (3, :a) (3, :b) ```...

**Summary** This attempts to address #512, which prevents kernels operating on vectors of unequal length from working properly. **Proposed changes** * ... Remove the check on input dimensions when the...

I would suggest either changing this, or perhaps better, skipping it all together. https://github.com/JuliaGaussianProcesses/KernelFunctions.jl/blob/ef6d4591b36194fca069d8bc7ae8c1e2ee288080/src/utils.jl#L194 I'm developing kernels that work on pairs of `Vector{Float64}`, where the kernel essentially sums over all...

**Summary** This is a minimal implementation to enable the simulation of gradients (and higher order derivatives) of GPs (see also https://github.com/JuliaGaussianProcesses/KernelFunctions.jl/issues/504) **Proposed changes** For a covariance kernel k of GP...

**Summary** The idea is that ```julia reduce(hcat, ColVecs(X)) ``` should in principle be a no-op. **Proposed changes** Specialization of `reduce` on `typeof(hcat)` and `typeof(vcat)`. The drawback is more code. But...

This might be a controversial point, but I have a few times been surprised by the fact that `Kernel` is an abstract type. I would like to propose/ consider adding...