In [2]:

    

%autosave 0
%matplotlib inline
import os, sys
sys.path.insert(0, os.path.expanduser('~/work/Dropbox/tmp/20160929-bml-github'))
sys.path.insert(0, os.path.expanduser('~/work/git/github/pymc-devs/pymc3'))

from bmlingam.tests.test_cli import test_cli

test_cli(plot=True)


    

    















    




Autosave disabled
Temporary files for test_cli()
data_file: /tmp/tmp_mxm6nlz
model_file: /tmp/tmptoxmvoxd

Made artificial data and saved as /tmp/tmp_mxm6nlz.
---- Algorithm parameters ----
Number of MC samples: 100
Number of candidate models: 1800

---- Data ----
Data loaded from /tmp/tmp_mxm6nlz.
Data contains 100 samples.
Variable names: ['x1_dst' 'x2_src']

---- Inference for variables "x1_dst" and "x2_src" ----
Inferred  : x2_src -> x1_dst (posterior prob: 0.711, loglikelihood: -268.213)
(best_rev): x1_dst -> x2_src (posterior prob: 0.015, loglikelihood: -272.078)

Hyper parameters of the optimal model:
Causality      : var2 -> var1
Standardize    : True
subtract_mu_reg: False
fix_mu_zero    : True
prior_var_mu   : auto
prior_indvdl   : t
v_indvdl_1     : 0.000000
v_indvdl_2     : 0.200000
df_indvdl      : 8.000000
L_cov12/21     : 0.300000
dist_noise     : gg
beta_noise     : 0.750000

Hyper parameters of the reverse optimal model:
Causality      : var1 -> var2
Standardize    : True
subtract_mu_reg: False
fix_mu_zero    : True
prior_var_mu   : auto
prior_indvdl   : t
v_indvdl_1     : 0.200000
v_indvdl_2     : 0.000000
df_indvdl      : 8.000000
L_cov12/21     : -0.500000
dist_noise     : gg
beta_noise     : 0.750000

Optimal model was saved as /tmp/tmptoxmvoxd.
/tmp/tmptoxmvoxd
---- Variables x1_dst and x2_src ----
Inferred causality   : x2_src -> x1_dst
Posterior mean       : 0.707533
95% Credible interval: (0.336794, 1.069183)

True True






    













    




Posterior samples was saved as /tmp/tmptoxmvoxd.

In [ ]: