Information Flows in Causal Networks

In [2]:

    

from causalinfo import *
from numpy import log2
from numpy.testing import assert_allclose
# You only need this if you want to draw pretty pictures of the Networksa
from nxpd import draw, nxpdParams
nxpdParams['show'] = 'ipynb'


    

In [3]:

    

w, x, y, z = make_variables("W X Y Z", 2)
wdist = UniformDist(w)


    

In [4]:

    

eq1 = Equation('BR', [w], [x, y], equations.branch_same_)
eq2 = Equation('XOR', [x, y], [z], equations.xor_)


    

In [5]:

    

# Build the graph
eg3 = CausalGraph([eq1, eq2])
draw(eg3.full_network)


    

    
Out[5]:

In [6]:

    

eg3


    

    
Out[6]:





<causalinfo.graph.CausalGraph at 0x115fc3d90>

In [7]:

    

m_eg3 = MeasureCause(eg3, wdist)

# See the table on p29a
assert m_eg3.mutual_info(x, y) == 1
assert m_eg3.mutual_info(x, y, w) == 0
assert m_eg3.mutual_info(w, z, y) == 0

assert m_eg3.causal_flow(x, y) == 0
assert m_eg3.causal_flow(x, y, w) == 0
assert m_eg3.causal_flow(w, z, y) == 1


    

In [8]:

    

def copy_first_(i1, i2, o1):
        o1[i1] = 1.0

eq2 = Equation('COPY_FIRST', [x, y], [z], copy_first_)


    

In [9]:

    

eg51 = CausalGraph([eq1, eq2])
draw(eg51.full_network)


    

    
Out[9]:

In [10]:

    

m_eg51 = MeasureCause(eg51, wdist)

# See paragraph at top of page 30
assert m_eg51.mutual_info(x, z, y) == 0
assert m_eg51.causal_flow(x, z, y) == 1
assert m_eg51.causal_flow(x, z) == 1


    

In [11]:

    

def random_sometimes_(i1, i2, o1):
    if i1 != i2:
        o1[:] = .5
    else:
        equations.xor_(i1, i2, o1)

eq2 = Equation('RAND', [x, y], [z], random_sometimes_)


    

In [12]:

    

eg52 = CausalGraph([eq1, eq2])
draw(eg52.full_network)


    

    
Out[12]:

In [13]:

    

m_eg52 = MeasureCause(eg52, wdist)

# See pg 30
expected = 3.0 / 4.0 * log2(4.0 / 3.0)
assert_allclose(m_eg52.causal_flow(x, z, y), expected)