Python generate_data_vector Examples

Example #1

#Example script to launch fit 1 dimensional mixture model. 


#Add the toolbox to path
import os 
import sys
toolbox_path = "../code" 
sys.path.append(os.path.join(os.path.abspath(toolbox_path))) 
#Add the toolbox to path


#Generate some data
generate_new_data=1
if generate_new_data==1:
    from generate_data_vector2 import generate_data_vector
    data_vector=generate_data_vector(N_comps=3, N_samples=5000 ,means=[ 0., 5. ,-5.], variances= [1.,2.,2.], mixing=[0.9, 0.09, 0.01])
else:
    import scipy.io as sio
    sio.loadmat('data_vector.mat')
#Generate some data
    

#Define options for the mixture model fit
Inference ='Variational Bayes'#'Method of moments'#'Variational Bayes'  #'Method of moments' OR 'Maximum Likelihood' OR 'Variational Bayes' ML NOT INCLUDED YET
Number_of_Components=3
Components_Model=['Gauss','Gamma','-Gamma'] #Each component can be Gauss, Gamma, InvGamma, -Gamma, -InvGamma
init_params=[0.,1.,6.,1.,-6.,1.]
init_pi=np.ones(3);
init_pi=np.divide(init_pi,3)
maxits=300
tol=0.00000001

Example #2

File: Mix_Mod_example.py Project: zahfaz/One_Dim_Mixture_Models

#Example script to launch fit 1 dimensional mixture model.

#Add the toolbox to path
import os
import sys

toolbox_path = "../code"
sys.path.append(os.path.join(os.path.abspath(toolbox_path)))
#Add the toolbox to path

#Generate some data
generate_new_data = 1
if generate_new_data == 1:
    from generate_data_vector2 import generate_data_vector
    data_vector = generate_data_vector(3, 50000, [0, 5, -5], [1, 1, 1],
                                       [0.8, 0.1, 0.1])
else:
    import scipy.io as sio
    sio.loadmat('data_vector.mat')
#Generate some data

#Define options for the mixture model fit
Inference = 'Variational Bayes'  #'Method of moments'#'Variational Bayes'  #'Method of moments' OR 'Maximum Likelihood' OR 'Variational Bayes' ML NOT INCLUDED YET
Number_of_Components = 3
Components_Model = [
    'Gauss', 'InvGamma', '-InvGamma'
]  #Each component can be Gauss, Gamma, InvGamma, -Gamma, -InvGamma
init_params = [0, 1, 5, 2, -5, 2]
maxits = 300
tol = 0.00000001
opts = {

Example #3

#Add the toolbox to path
import time

import os
import sys
import numpy as np

toolbox_path = "../code"
sys.path.append(os.path.join(os.path.abspath(toolbox_path)))
#Add the toolbox to path

#Generate some data
generate_new_data = 1
if generate_new_data == 1:
    from generate_data_vector2 import generate_data_vector
    data_vector = generate_data_vector(3, 50000, [0, 3, -3], [1, 1, 1],
                                       [0.9, 0.05, 0.05])
else:
    import scipy.io as sio
    sio.loadmat('data_vector.mat')
#Generate some data

#Define options for the mixture model fit
Inferences_possibilities = [
    'Method of moments', 'Maximum Likelihood', 'Variational Bayes'
]
Number_of_Components = 3
#Each component can be Gauss, Gamma, InvGamma, -Gamma, -InvGamma
init_params = [0, 1, 5, 2, -5, 2]
init_pi = np.ones(3)
init_pi = np.divide(init_pi, 3)
maxits = 300