def generate_model(p=30, p_zero=0.7, type=None):
DEBUG = True
'''
this function creates a model in order to generate
data according to it's distribution.
and in order to compare the approximated model to the original
distribution
'''
if not type:
adjmat = random_graph(p, p_zero)
if type == 'grid':
adjmat = grid_graph(p)
adjmat.nodes()
adjmat.nodes_iter()
gen = generate(adj_mat=adjmat)
return gen
def random_graph(graph=None, size=10, cluster=True, layout=None):
"""Return a random graph (models.py Graph class).
This was my original test graph function. It makes a 2D
non-periodic lattice, adds some random noise to it, and returns it.
"""
if graph is None:
g = grid_graph(dim=[size, size])
g = nx_relabel_nodes(g, dict((n, i) for i, n in enumerate(g.nodes())))
if layout:
layout = nx.drawing.nx_pydot.pydot_layout(g)
# layout = nx.drawing.layout.shell_layout(g)
if cluster:
clusterGraph(g, p=0.05)
# Set weights and make the Graph object.
for a, b, d in g.edges(data=True):
d["weight"] = -10
from .old.models import Graph
G = Graph.fromNetworkX(g, coords=layout, defaultweight=1)
return G
def random_graph(graph=None, size=10, cluster=True, layout=None):
"""Return a random graph (models.py Graph class).
This was my original test graph function. It makes a 2D
non-periodic lattice, adds some random noise to it, and returns it.
"""
if graph is None:
g = grid_graph(dim=[size, size])
g = nx_relabel_nodes(g, dict((n, i) for i, n in enumerate(g.nodes())))
if layout:
layout = nx.drawing.nx_pydot.pydot_layout(g)
# layout = nx.drawing.layout.shell_layout(g)
if cluster:
clusterGraph(g, p=.05)
# Set weights and make the Graph object.
for a, b, d in g.edges(data=True):
d['weight'] = -10
from .old.models import Graph
G = Graph.fromNetworkX(g, coords=layout, defaultweight=1)
return G
def __init__(self, lattice=ngc.grid_graph( dim=[N,N] ),
data=zeros((N,N)), tau_x = 1, tau_y = 1, phi = 0.1):
# sanity test
if lattice.number_of_nodes() != data.size:
raise Exception('data and lattice sizes do not match',
'%d vs %d' % (data.size, lattice.number_of_nodes()) );
self.num_nodes = lattice.number_of_nodes();
self.phi = phi;
self.tau_x = 1;
self.tau_y = 1;
#just in case the input decides to give us weights
for e in lattice.edges_iter():
if not lattice.get_edge_data(e[0],e[1]) :
#setting lattice
lattice.edge[e[0]][e[1]] = {'weight':phi};
lattice.edge[e[1]][e[0]] = {'weight':phi};
else:
#keep the data
pass;
self.lattice , self.data = lattice, data;
# convert the lattice into a GMRF precision matrix
self.Lambda = zeros(self.num_nodes,self.num_nodes);
#set up the grid and the data
#@stochastic(dtype=float)
#@def X
# this v is a tuple index of the grid
self.Y = [ Normal('Y_'+str(v), mu=0.5, tau=Sigma_Y**-1,
value=data[v], observed = True )
for v in lattice.nodes_iter() ];
MCMC.__init__(self, [self.Y])
# Richard Darst, August 2011
import numpy
import networkx
from networkx.generators.classic import grid_graph
import util
import pcd
g = grid_graph([20,20], periodic=True)
for a,b,d in g.edges(data=True):
d['weight'] = -1
G = pcd.Graph.fromNetworkX(g, defaultweight=1)
print (numpy.sum(G.imatrix) - numpy.sum(G.imatrix.diagonal()))/(400**2-400.)
#exit()
assert numpy.all(G.imatrix - G.imatrix.T == 0)
MR = pcd.MultiResolution()
MR.run([G]*5, gammas=dict(low=.1, high=100, density=10))
MR.write('tmp-lattice.txt')
#MR.viz()
# Richard Darst, August 2011
import numpy
import networkx
from networkx.generators.classic import grid_graph
import util
import pcd
g = grid_graph([20, 20], periodic=True)
for a, b, d in g.edges(data=True):
d['weight'] = -1
G = pcd.Graph.fromNetworkX(g, defaultweight=1)
print(numpy.sum(G.imatrix) - numpy.sum(G.imatrix.diagonal())) / (400**2 - 400.)
#exit()
assert numpy.all(G.imatrix - G.imatrix.T == 0)
MR = pcd.MultiResolution()
MR.run([G] * 5, gammas=dict(low=.1, high=100, density=10))
MR.write('tmp-lattice.txt')
#MR.viz()
