Stack Overflow
Stack Overflow

Questions tagged [mixture-model]

In statistics, a mixture model is a probabilistic model that is formed by a combination of probabilistic models. For example, the Gaussian mixture model is the weighted sum of Gaussian distributions. Mixture models are mainly used in probabilistic clustering of data and modelling of unknown distributional shapes.

222 questions
1
vote
3 answers

Gaussian Mixture Model Components

I am trying to fit my data to Gaussian Mixture Model using matlab , but the problem is that I can't determine the optimal number of components to do this , Can any body help !!! Also if there are already build functions to get that optimal number…
Ameer Sameer
  • 19
  • 1
  • 2
0
votes
1 answer

Object modelling using openCV GMM class

BACKDROP: In openCV video processing, I have segmented foreground as n number of blobs. Question: OpenCV 2.3 provides a class called GMM. How can I model a blob's pixels using a GMM. I am new to Gaussian mixture modelling. I can undestand its…
user1036908
  • 831
  • 1
  • 12
  • 16
0
votes
0 answers

How do I set a different prior distribution for each Numpyro Component

I'm building mixture model in Numpyro, and would like to set each component with a different prior. So far I have this where n == the number of components in the mixture. def model(data, n): weights = numpyro.sample('weights',…
caceves
  • 23
  • 6
0
votes
0 answers

Expectation Step in Gaussian Mixture Model for Matrix Data Not Producing Proper Posterior Probabilities

I'm working on implementing a Gaussian Mixture Model (GMM) for three-way data (i.e., a set of matrices) in R. The GMM is being estimated using the Expectation-Maximization (EM) algorithm. However, I'm encountering an issue during the Expectation (E)…
Omarov Alen
  • 25
  • 3
0
votes
1 answer

How to manually define intial parameters for scikit GMM

I am using scikit's General Mixture Model to fit some data, but I would like to define the initial centroids of each cluster manually. The scikit learn documentation states: There is a choice of four initialization methods (as well as inputting…
Thai Pro
  • 11
  • 1
0
votes
0 answers

How to find the joint probability using copula?

Let X1 ~ F2 = p1F10 + (1-p1)F11 X1 ~ F2 = p1F20 + (1-p2)F21 and P(X1 <= x1, X2 <= x2) = C(F1i(x1), F2j(x2)) (C is the same copula for all i,j=0,1) I know the exact p1, p2, all Fij and C. I think P(X1 from F10, X2 from F20) is not equal to p1*p2 but…
황서화
  • 9
  • 2
0
votes
0 answers

Fit a bimodal histogram with a mixture Model

I am attempting to fit with Python a histogram which presents a bimodal distribution. For now, I have tested to use the Gaussian Mixture Model (GMM) from Scikit-Learn, but I want to try with different combinations of distributions that are not…
Jack
  • 35
  • 1
  • 6
0
votes
0 answers

determine if data is uni- or bimodal

I have a dataset where many but not all of my data seems to have a bivariate distribution. I can perform a mixture model for each group in the dataset, but I'd like to know how to test if a univariate model is better fit. dfA <- filter(df, group ==…
tnt
  • 1,149
  • 14
  • 24
0
votes
0 answers

Training `tfd.MixtureSameFamily` gets 'NaN'

I want to train a mixture density model using tfd.MixtureSameFamily. But after several thousand epochs of training, the result gets NaN. Here's full functioning code to replicate this. import section import tensorflow as tf import…
David H. J.
  • 340
  • 2
  • 12
0
votes
0 answers

Error with overlaying histogram with multiple beta distribution in ggplot2

I am trying to evaluate the fit of my mixture model by overlaying the histogram of the original data with three beta distributions. I am using ggplot2 for visualization, but I am facing issues with the overlay. The histogram of the original data is…
hollyjolly
  • 115
  • 11
0
votes
0 answers

Dirichlet Process manually

I am implementing the Dirichlet Mixture Model using the EM algorithm in R, but am experiencing issues with the results. I generated two binomial distributions with fractions of (70%, 30%) and means of (0.05, 0.18), and trimmed 5% of the data set…
hollyjolly
  • 115
  • 11
0
votes
0 answers

Split chronological data in k groups to optimize k linear regressions

I have X and Y data arranged in chronological order. I want to find the k-1 bounds that divide my data into k clusters so that each cluster optimizes a linear regression Yk = ak.Xk + bk I tried an iterative search for the bounds but without much…
bndgte
  • 1
  • 1
0
votes
1 answer

EM algorithm with Beta Distribution in R

I would like to write R code to build the dirichlet mixture model. The loglikelihood I used for the beta distribution is as below: (,)=(−1)ln¯+(−1)ln(1−)¯+lnΓ(+)−lnΓ()−lnΓ() and I need help for initialising parameters (alpha, beta and setting…
hollyjolly
  • 115
  • 11
0
votes
0 answers

How to evaluate probabilities of samples in Gaussian Mixture Model

I've created a Gaussian Mixture Model in Python with 11 components based on 8-dimensional data (I picked 11 components because that's what minimized the BIC score). I now have a test sample of data (50 samples of 8-dimensional data), and I want to…
curious_cosmo
  • 1,184
  • 1
  • 18
  • 36
0
votes
0 answers

Weighted mixture model of two distributions where weight depends on the value of the distribution?

I'm trying to replicate the precipitation mixture model from this paper: http://dx.doi.org/10.1029/2006WR005308 f(r) is the gamma PDF, g(r) is the generalized Pareto PDF, and w(r) is the weighting function, which depends on the value r being…
cbowers
  • 137
  • 8
1 2 3
…
14 15