public void WriteRecords(string filePath, Dictionary <string, object> extra = null)
{
MatlabWriter writer = new MatlabWriter(filePath);
WriteGammaList(writer, InPrecision, "inShape", "inRate");
WriteGammaList(writer, OutPrecision, "outShape", "outRate");
// MatlabWritter cannot write boolean
writer.Write("consultOracle", MatrixUtils.ToDouble(ConsultOracle));
Matrix uncertaintyMat = MatrixUtils.StackColumnsIgnoreNull(Uncertainty);
// write a Matrix
writer.Write("uncertainty", uncertaintyMat);
WriteGammaList(writer, OracleOut, "oraOutShape", "oraOutRate");
if (extra != null)
{
foreach (var kv in extra)
{
writer.Write(kv.Key, kv.Value);
}
}
writer.Dispose();
}
public static void TestWritingMat()
{
string p = Config.PathToSavedFile("test_save_mat.mat");
MatlabWriter w = new MatlabWriter(p);
w.Write("b", 2);
double[] arr = new double[] { 1, 2, 3 };
w.Write("arr", arr);
Vector vec = Vector.FromArray(new double[] { 4, 5, 6, 7 });
w.Write("vec", vec);
List <double> list = new List <double>(arr);
w.Write("list", list);
Matrix m = Matrix.Parse("1 2\n 3 4");
w.Write("m", m);
long time = 1329L;
w.Write("longNum", time);
List <Matrix> mats = new List <Matrix>();
mats.Add(Matrix.IdentityScaledBy(2, 3.0));
mats.Add(Matrix.IdentityScaledBy(3, 4.0));
w.Write("list_mats", mats);
w.Dispose();
}
public static void ToLogisticSerializeToMat(string file,
List <Tuple <Beta, Gaussian, Beta> > msgs,
Dictionary <string, object> extra)
{
int n = msgs.Count;
double[] outBetaA = new double[n];
double[] outBetaB = new double[n];
double[] inNormalMeans = new double[n];
double[] inNormalVariances = new double[n];
// alpha parameters
double[] inBetaA = new double[n];
double[] inBetaB = new double[n];
for (int i = 0; i < msgs.Count; i++)
{
Tuple <Beta, Gaussian, Beta> pairI = msgs[i];
Beta toBeta = pairI.Item1;
Gaussian fromX = pairI.Item2;
Beta fromLogistic = pairI.Item3;
fromX.GetMeanAndVariance(out inNormalMeans[i], out inNormalVariances[i]);
outBetaA[i] = toBeta.TrueCount;
outBetaB[i] = toBeta.FalseCount;
double alpha = fromLogistic.TrueCount;
double beta = fromLogistic.FalseCount;
inBetaA[i] = alpha;
inBetaB[i] = beta;
}
// write to .mat file
MatlabWriter matWriter = new MatlabWriter(file);
matWriter.Write("outBetaA", outBetaA);
matWriter.Write("outBetaB", outBetaB);
matWriter.Write("inNormalMeans", inNormalMeans);
matWriter.Write("inNormalVariances", inNormalVariances);
matWriter.Write("inBetaA", inBetaA);
matWriter.Write("inBetaB", inBetaB);
if (extra != null)
{
foreach (var kv in extra)
{
matWriter.Write(kv.Key, kv.Value);
}
}
matWriter.Dispose();
}
public void RecordInferNETTime()
{
/**Records time by infer.net*/
/**
* Only one W just like in Ali's paper.
* In practice, we typically observe multiple sets of observations
* where we want to do inference on the same model with the same
* parameter.
*/
Rand.Restart(init_fixed_seed);
Vector w = Vector.Zero(d);
Rand.Normal(Vector.Zero(d), PositiveDefiniteMatrix.Identity(d), w);
// Create the Logistic operator instance only one because we want to use the same
// one after a new problem (new seed).
// stopwatch for measuring inference time for each problem
Stopwatch watch = new Stopwatch();
Type logisticOp = typeof(LogisticOp2);
LogisticOp2.Watch = watch;
List <long> allInferTimes = new List <long>();
var allPosteriors = new List <VectorGaussian>();
LogisticOp2.IsCollectLogisticMessages = false;
LogisticOp2.IsCollectProjMsgs = false;
LogisticOp2.IsCollectXMessages = false;
for (int seed = seed_from; seed <= seed_to; seed++)
{
Rand.Restart(seed);
double b = 0;
// combine the bias term into W
Vector[] X;
bool[] Y;
LogisticRegression.GenData(n, w, b, out X, out Y, seed);
Console.Write("Y: ");
StringUtils.PrintArray(Y);
VectorGaussian wPost;
// start the watch
watch.Restart();
LogisticRegression.InferCoefficientsNoBias(X, Y, out wPost, epIter, logisticOp);
// stop the watch
long inferenceTime = watch.ElapsedMilliseconds;
allInferTimes.Add(inferenceTime);
allPosteriors.Add(wPost);
//print
Console.WriteLine("n: {0}", n);
Console.WriteLine("d: {0}", d);
int t = Y.Sum(o => o ? 1 : 0);
Console.WriteLine("number of true: {0}", t);
Console.WriteLine("True bias: {0}", b);
// Vector meanW = wPost.GetMean();
Console.WriteLine("True w: {0}", w);
Console.WriteLine("Inferred w: ");
Console.WriteLine(wPost);
}
string fnameM = string.Format("rec_dnet_n{0}_logistic_iter{1}_sf{2}_st{3}.mat",
n, epIter, seed_from, seed_to);
string recordPathM = Config.PathToSavedFile(fnameM);
MatlabWriter writer = new MatlabWriter(recordPathM);
writer.Write("allInferTimes", MatrixUtils.ToDouble(allInferTimes));
Vector[] postMeans = allPosteriors.Select(vg => vg.GetMean()).ToArray();
Matrix[] postCovs = allPosteriors.Select(vg => vg.GetVariance()).ToArray();
writer.Write("postMeans", postMeans);
writer.Write("postCovs", postCovs);
writer.Write("dim", d);
writer.Write("n", n);
writer.Write("epIter", epIter);
writer.Write("seed_from", seed_from);
writer.Write("seed_to", seed_to);
writer.Write("init_fixed_seed", init_fixed_seed);
writer.Dispose();
}
