Gmm.FoleyGmm

FoleyGmm.py.

adapted from:

https://towardsdatascience.com/gaussian-mixture-modelling-gmm-833c88587c7f https://www.kaggle.com/dfoly1/gaussian-mixture-model

Copyright (c) 2020, SAXS Team, KEK-PF

class GMM(C, n_runs)

Bases: object

Gaussian Mixture Model

Parameters:

• k (int , number of gaussian distributions)

• seed (int, will be randomly set if None)

• max_iter (int, number of iterations to run algorithm, default: 200)

centroids

Type:

array, k, number_features

cluster_labels

Type:

label for each data point

calculate_mean_covariance(X, prediction)

Calculate means and covariance of different

clusters from k-means prediction

Parameters:

prediction: cluster labels from k-means X: N*d numpy array data points

Returns:

intial_means: for E-step of EM algorithm intial_cov: for E-step of EM algorithm

fit(X)

Compute the E-step and M-step and

Calculates the lowerbound

Parameters:

X: (N x d), data

Returns:

instance of GMM

get_params()

predict(X)

Returns predicted labels using Bayes Rule to Calculate the posterior distribution

Parameters:

X: ?*d numpy array

Returns:

labels: predicted cluster based on highest responsibility gamma.

predict_proba(X)

Returns predicted labels

Parameters:

X: N*d numpy array

Returns:

labels: predicted cluster based on highest responsibility gamma.