ranjanan

using OrdinaryDiffEq
using Sundials
using Plots
using BenchmarkTools

const N = 40  # Number of heated units

const Cu = N == 1 ? [2e7] : (ones(N) .+ range(0,1.348,length=N))*1e7 # "Heat capacity of heated units";
const Cd = 2e6*N # "Heat capacity of distribution circuit";
const Gh = 200 # "Thermal conductance of heating elements";

ranjanan

using OrdinaryDiffEq
using DiffEqFlux
using Flux
using Plots

#=function f(du, u, p, t)
    du[1] = -p[1]*u[1] + p[2]*u[2]*u[3]
    du[2] = p[1]*u[1] - p[2]*u[2]*u[3] - p[3]*u[2]*u[2]
    du[3] = p[3]*u[2]*u[2]
end=#

ranjanan

# Add this first: https://github.com/ranjanan/MonteCarloIntegration.jl
using Cubature
# using Cuba
using MonteCarloIntegration

# include("vegas.jl")

function koopman(g,prob,u0,p,args...;kwargs...)
    g(solve(remake(prob,u0=u0,p=p),args...;kwargs...))
end

ranjanan

import DecisionTree
using ScikitLearn
@sk_import tree: DecisionTreeClassifier
import PyCall: PyObject
using JuliaDB

function load_data()

    isfile("balance.data") || 
        download("https://archive.ics.uci.edu/ml/machine-learning-databases/balance-scale/balance-scale.data", 

ranjanan

ranjanan

Running tests:
	From worker 3:	    testing Distributions.CategoricalDirectSampler
	From worker 3:	    testing Distributions.AliasTable
	From worker 3:	    testing Distributions.BinomialGeomSampler
	From worker 3:	    testing Distributions.BinomialTPESampler
	From worker 3:	    testing Distributions.BinomialPolySampler
	From worker 2:	    [Discrete]
	From worker 5:	    testing Distributions.Categorical(K=2, p=[0.5,0.5])
	From worker 2:	    ------------
	From worker 2:	    testing Bernoulli()

ranjanan

First, 

git clone https://github.com/ranjanan/MT-Workloads.git

Then cd into ALS/

On the julia REPL, 

include("ALS.jl")
main()