def transform_point(point, transform):
"""transform a point"""
hpoint = Vec(point)
hpoint.append(1.0)
hres = transform.mmul(hpoint)
res = vector.vector(hres[0:-1]) / hres[-1]
return res
def transform_point(point, transform):
"""transform a point"""
hpoint = Vec(point)
hpoint.append(1.0)
hres = transform.mmul(hpoint)
res = vector.vector(hres[0:-1]) / hres[-1]
return res
def transform(self, t):
"""returns a new configuration, which is this one transformed by matrix t"""
newmap = {}
for v in self.map:
p = self.map[v]
ph = Vec(p)
ph.append(1.0)
ph = t.mmul(ph)
p = vector.vector(ph[0:-1]) / ph[-1]
newmap[v] = p
return Configuration(newmap)
def transform(self, t):
"""returns a new configuration, which is this one transformed by matrix t"""
newmap = {}
for v in self.map:
p = self.map[v]
ph = Vec(p)
ph.append(1.0)
ph = t.mmul(ph)
p = vector.vector(ph[0:-1]) / ph[-1]
newmap[v] = p
return Configuration(newmap)
def perp2D(v):
w = Vec(v)
w[0] = -v[1]
w[1] = v[0]
return w
import hmm_forward from matfunc import Table, Vec, Matrix ## example 10.2 from Li Hang's book (Statistical Learning Methods), pg. 177 a1 = Table([0.5, 0.2, 0.3]) a2 = Table([0.3, 0.5, 0.2]) a3 = Table([0.2, 0.3, 0.5]) A = Matrix([a1, a2, a3]) #print "matrix A is:\n", A b1 = ([0.5, 0.5]) b2 = ([0.4, 0.6]) b3 = ([0.7, 0.3]) B = Matrix([b1, b2, b3]) #print "matrix B is:\n", B pi = Vec([0.2, 0.4, 0.4]) T = 3 O = Table([0, 1, 0]) # red , white, red print "The prob is:\n", hmm_forward.forward(A, B, pi, O)
