I am refactoring my source code, which previously used List<T>, to make it memoryless by introducing IEnumerable<T> and yield return so that my source code can handle bigdata.

Suppose I have the following function, which is already refactored in favor of IEnumerable<T>:

    static class MyClass
    {
        public static (IEnumerable<double> XFit, IEnumerable<double> YFit, IEnumerable<double> MinimizingPoint) 
            Fit(IEnumerable<double> xData, IEnumerable<double> yData, int minIterations = 100000)
        {
            var xDataDense = new DenseVector(xData.ToArray());
            var yDataDense = new DenseVector(yData.ToArray());

            Vector<double> Model(Vector<double> parameters, Vector<double> x)
            {
                var y = CreateVector.Dense<double>(x.Count);
                for (int i = 0; i < x.Count; i++)
                {
                    y[i] = parameters[0] * Math.Exp(parameters[1] * x[i]);
                }
                return y;
            }

            var start = new DenseVector(new double[] { 1.0, 0.1 });
            var objective = ObjectiveFunction.NonlinearModel(Model, xDataDense, yDataDense);
            var solver = new LevenbergMarquardtMinimizer(maximumIterations: minIterations);
            var result = solver.FindMinimum(objective, start);

            var points = result.MinimizedValues.ToArray();
            var minimizing = result.MinimizingPoint.ToArray();

            return (xData, points, minimizing);
        }
    }

This function is called in another function:

    class MyClass2
    {
public static (double A0, double Tau0) GetA0Tau0ForVec3(IEnumerable<Vec3MemoryLess> rEndVecs, int numLags = 1000, bool isLog = false)
        {
            var autocorr = TimeSeriesAnalysisMemoryLess.GetAutoCorrelationPoints(rEndVecs, numLags);

            var TauListX = autocorr.lags;//Error: doesn't work
            var AutoCorrelationListY = autocorr.;//Error: doesn't work

            var fittedCurve = NonLinearRegressionCurveFittingMemoryLess.Fit(autocorr.lags, autocorr.autocorrs);

            IEnumerable<double> xExpDecayList = fittedCurve.XFit;
            IEnumerable<double> yExpDecayList = fittedCurve.YFit;
            IEnumerable<double> results = fittedCurve.MinimizingPoint;

            double A0Value = 0;

            if (isLog)
            {
                A0Value = Math.Exp(results.First());
            }
            else
            {
                A0Value = results.First();
            }

            double Tau0Value = 1.0 / ((-1.0) * results.Skip(1).First());

            return (A0Value, Tau0Value);
        }
    }

How can I apply yield return in these functions?

There is also a problem with tuple splitting in the second function.