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
0
votes
1 answer

How to create initial value estimation using expectation-maximization algorithm in SPSS?

I would like to use EM algorithm to create initial values in spss. I have used missing value analysis in spss and also selected EM and also I chose 500 times as the maximum number of iterations but after running the spss I just get one set of data!…
Ava
  • 67
  • 1
  • 9
0
votes
0 answers

Transparent image after applying expectation maximization

I have applied the expectation maximization on my image. But when I run the executable file, it shows me a transparent screen. Can anyone please tell me that what is the problem in my code?? Thanks Here is my code #include…
Rishav Bhola
  • 1
  • 1
0
votes
0 answers

Expectation-Maximization in Weka for custom objects

I'm planning to use the Java Weka library's EM algorithm in order to assign probabilities to objects to be in a certain cluster and then, work with these probabilities. Furthermore, the properties of those objects will be loaded from a database, so…
aloplop85
  • 892
  • 3
  • 16
  • 40
0
votes
0 answers

Error within function using solve() in R

With this program below, I will get the error: Error in solve.default(sig[!pick.miss, !pick.miss]) : 'a' is 0-diml I want to use EM algorthim to impute the mising values.This function can work if the missing values is small(from 5% to 20% of the…
angelia
  • 1
0
votes
1 answer

Using Gaussian mixture model for labelled and unlabelled data in scikit learn?

I have data out of which some of it is labelled and some of it is not labelled. I need to apply Expectation maximization algorithm. Is it possible to apply it in Sci-kit learn? Any example of labelled + unlabelled instances would be helpful.
user2793286
  • 179
  • 2
  • 14
0
votes
1 answer

Getting EM clusterer output programmatically in Weka

When using Kmeans in Weka, one can call getAssignments() on the resulting output of the model to get the cluster assignment for each given instance. I am using EM clustering on weka. I want to get output like weka's explorer as bellow or is there…
Arash m
  • 99
  • 2
  • 12
0
votes
0 answers

EM-Algorithm with parameter restrictions using R

Using any type of data, in my case data from three gamma distributions mixed, the goal is to parameterise the distributions' theta and the distribution weight alpha that are all 0< and sum to 1. How do I restrict the alphas to sum to one, where no…
BDL
  • 1
0
votes
0 answers

One line exaggerated execution time

I'm developing a program with OpenCV in C++. The program it's working fine. The line of code that I've put below normally needs some miliseconds to execute but sometimes the same line needs 5, 10 or even 50 seconds. clustEM.trainE(resP.colRange(0,…
Ricardo
  • 300
  • 1
  • 3
  • 12
0
votes
0 answers

Use C++ armadillo Expectation Maximization for Gaussian mixture model

I am trying to use armadillo's expectation maximization maximization gmm_diag class, but when I try to compile it I get "error gmm_diag was not declared in this scope". My code is as follows: #include #include #include…
Sisay Zinabu
  • 71
  • 2
  • 14
0
votes
0 answers

what is hidden(unobserved) data? and what is hidden data in GMM?

I am studying EM algorithm and GMM together I don't understand the EM algorithm as described in Wikipedia: The EM algorithm is used to find the maximum likelihood parameters of a statistical model in cases where the equations cannot be solved…
user2874612
  • 39
  • 1
  • 4
0
votes
1 answer

Enhancing the performance of Expectation Maximization

I am implementing expectation maximization (EM) in C++ to estimate the parameter of Gaussian mixture model. The EM is very slow to converge - is there a technique to converge log likelihood in a fast way?
kahsay kalayu
  • 309
  • 5
  • 16
0
votes
2 answers

Benchmarking EM Soft Clustering vs K-Means?

I have two implementations one is K-Means and the other is EM doing soft clustering. But I do not know how to validate them in terms of accuracy. i.e. which one performs better by retrieving better clusters. My assumption is that because EM is doing…
Walter white
  • 173
  • 2
  • 11
0
votes
1 answer

A framework for comparing the time performance of Expectation Maximization

I have my own implementation of the Expectation Maximization (EM) algorithm based on this paper, and I would like to compare this with the performance of another implementation. For the tests, I am using k centroids with 1 Gb of txt data and I am…
Walter white
  • 173
  • 2
  • 11
0
votes
1 answer

How to compute the variances in Expectation Maximization with n dimensions?

I have been reviewing Expectation Maximization (EM) in research papers such as this one: http://pdf.aminer.org/000/221/588/fuzzy_k_means_clustering_with_crisp_regions.pdf I have some doubts that I have not figured it out. For example, what would…
Walter white
  • 173
  • 2
  • 11
0
votes
1 answer

EM clustering based background foreground segmentation in OPENCV

I tried to perform EM based back ground foreground segmentation using a code below...which I also found in Stackoverflow....But seems there is some error somewhere as I dont ever see the second printf statement to get executed... . basically it is…
noviceS
  • 1
  • 1
1 2 3
…
8 9