Stack Overflow
Stack Overflow

Questions tagged [expectation-maximization]

Expectation Maximization (often abbreviated EM) is an iterative algorithm that can be used for maximum likelihood estimation in the presence of missing data or hidden variables.

125 questions
1
vote
0 answers

Do Expectation Maximization always have to converge after a finite number of iterations if the implementation is correct?

I'm trying to understand how this could be possible, I know EM algorithm has the property to increase the likelihood for each step. However, this does not imply convergence. My question is, if the implementation IS proper and correct, EM will always…
tu_1329
  • 25
  • 6
1
vote
1 answer

Pgmpy: expectation maximization for bayesian networks parameter learning with missing data

I'm trying to use the PGMPY package for python to learn the parameters of a bayesian network. If I understand expectation maximization correctly, it should be able to deal with missing values. I am currently experimenting with a 3 variable BN, where…
TdV
  • 11
  • 1
1
vote
1 answer

Expectation Maximization using a Poisson likelihood function

I am trying to apply the expectation-maximization algorithm to estimate missing count data but all the packages in R, such as missMethods, assume a multivariate Gaussian distribution. How would I apply the expectation-maximization algorithm to…
Constantin
  • 132
  • 9
1
vote
1 answer

multivariate_normal.pdf() returns zeros?

I am trying to use Expectation Maximization (EM) to classify my data. But due to my limited experience with Python and algorithms, when I writing the E-step of the algorithm, the method multivariate_normal.pdf() returns zeros and I have no idea why…
Rick
  • 61
  • 7
1
vote
2 answers

is statsmodels' MLEModel class using Expectation Maximization for fitting?

I am building a custom stat space model using statsmodels' MLEModel class, and fitting the unknown parameters with the .fit() method. I thought that it was using Expectation Maximization, but I am not sure of that and I cannot find any hint to this…
deltasun
  • 314
  • 3
  • 11
1
vote
1 answer

R Error in if (nrow(emobj$Mu) != nclass || ncol(emobj$Mu) != p || nrow(emobj$LTSigma) != : missing value where TRUE/FALSE needed

I am getting the following error while using the EMCluster library in R: Error in if (nrow(emobj$Mu) != nclass || ncol(emobj$Mu) != p || nrow(emobj$LTSigma) != : missing value where TRUE/FALSE needed This is the code I wrote: emcluster(iris[,-5],…
Shreyas Chaturvedi
  • 15
  • 3
1
vote
0 answers

How to check correctness of implemented expectation-maximization algorithm in r?

Good afternoon ! I asked this question many times without any reply or comment (even in cross validated)! Under R, I had implemented the expectation-maximization algorithm for gaussian-mixture model : k=3 # number of known clusters w_k=rep(1,k)/k…
Tou Mou
  • 1,270
  • 5
  • 16
1
vote
0 answers

Clustering Strings (text) with Kmeans/EM using Levenshtein Distance

I am trying to cluster strings using Kmeans/EM. I have a list of strings (about 70 strings) and I want to cluster them using Levenshtein similarity metric. So basically, I am trying to implement the clustering part in this research paper:…
Kushagra Bhatia
  • 113
  • 1
  • 8
1
vote
1 answer

Bounded Optimization using scipy Libraries

Let us suppose a loglikelihood function f(x, y, z) = prob(k0)* log((1-x)^(1-y)) + prob(k1)*log((1-x)^(1-z)) and there exists constraints such that possible values of x, y and z should lie between 0 and 1. The goal is to minimize the function and…
Rituraj Singh
  • 579
  • 1
  • 5
  • 16
1
vote
0 answers

c++ OpenCV how can I use EM algorithm on a superpixel?

I need to use the EM algorithm to split out every superpixel of my image in two groups using the Expectation Maximization algorithm of OpenCV. I have initialized the image and the mask of the i-th superpixel, I copied the pixels of my superpixel to…
Alberto Pierini
  • 11
  • 2
1
vote
1 answer

Gaussian Mixture Model - Singular matrix

During the calculation of a Gaussian Mixture Model I have to calculate the pdf() of the multivariate Gaussian distribution. I do this using the scipy multivariate_normal() method. Now during the calculations I encounter the situation that I get an…
2Obe
  • 3,570
  • 6
  • 30
  • 54
1
vote
1 answer

Expectation Maximization algorithm(Gaussian Mixture Model) : ValueError: the input matrix must be positive semidefinite

I am trying to implement Expectation Maximization algorithm(Gaussian Mixture Model) on a data set data=[[x,y],...]. I am using mv_norm.pdf(data, mean,cov) function to calculate cluster responsibilities. But after calculating new values of covariance…
Madhura Raut
  • 51
  • 1
  • 5
1
vote
0 answers

Fuzzy clustering in WEKA

I'm trying to perform fuzzy clustering using WEKA (both GUI and java code). In my data set I have two fields: id and string, so I would like to cluster by these strings and as an output get per each string, array with probabilities of belonging to…
Slavenya
  • 11
  • 3
1
vote
1 answer

Handling zero rows/columns in covariance matrix during em-algorithm

I tried to implement GMMs but I have a few problems during the em-algorithm. Let's say I've got 3D Samples (stat1, stat2, stat3) which I use to train the GMMs. One of my training sets for one of the GMMs has in nearly every sample a "0" for stat1.…
Evox33
  • 98
  • 1
  • 8
1
vote
1 answer

Speeding up guassian EM algorithm

My python code is as follows...it takes forever. There must be some numpy tricks I can use? The picture I am analyzing is tiny and in grayscale... def gaussian_probability(x,mean,standard_dev): termA = 1.0 /…
Chris
  • 28,822
  • 27
  • 83
  • 158
1 2
3
…
8 9