GaussianMixtureRegressions provides implementation of Gaussian mixture regression model1 in Julia.
julia> using GaussianMixtureRegressions
julia> n = 100;
julia> d = 3;
julia> w = [1., 2., 3.];
julia> b = -1.0;
julia> σ = 2.0;
julia> X = randn(n, d);
julia> y = X*w .+ b .+ randn(n)*σ;
julia> X = hcat(X, y);
julia> K = 3;
julia> model = fit(GMR{K}, X)
[ Info: Initializing GMM, 3 Gaussians diag covariance 4 dimensions using 100 data points
K-means converged with 5 iterations (objv = 496.2564299844352)
┌ Info: K-means with 100 data points using 5 iterations
└ 6.7 data points per parameter
GMR{3, MixtureModel{Multivariate, Continuous, FullNormal, Categorical{Float64, Vector{Float64}}}}(MixtureModel{FullNormal}(K = 3)
components[1] (prior = 0.3146): FullNormal(
dim: 4
μ: [0.26776175162691485, 0.2509249295364129, 0.9660834976335448, 3.532775397209546]
Σ: [0.7115861947660863 -0.1944775509058701 -0.20333074136580565 -0.13206921117259834; -0.1944775509058701 0.8514434620519014 0.011086817365474606 1.2602845902443185; -0.20333074136580565 0.011086817365474606 0.2348133924977131 0.3688123774069542; -0.13206921117259834 1.2602845902443185 0.3688123774069542 5.354242427805917]
)
...GaussianMixtureRegressions.save(model, "model.bson")
gmr = GaussianMixtureRegressions.load("model.bson")- Ghahramani, Z., & Jordan, M. (1994). Supervised learning from incomplete data via an EM approach. In J. Cowan, G. Tesauro, & J. Alspector (Eds.), Advances in neural information processing systems (Vol. 6, pp. 120–127). Morgan-Kaufmann.
-
Fabisch, A. (2021). gmr: Gaussian Mixture Regression. Journal of Open Source Software, 6(62), 3054. https://doi.org/10.21105/joss.03054