def compute_S(M):
n = 26
m = 2*(26*26*27)/2
X = zeros((m,(n*(n+1))/2))
Y = zeros(m)
r = 0
for i in range(n):
for j in range(n):
for k in range(j+1):
X[r][to_in(i,j)]=1.
X[r][to_in(i,k)]=-1.
Y[r] = M[i][j][k]
r+=1
X[r][to_in(i,j)]=-1.
X[r][to_in(i,k)]=1.
Y[r] = M[i][k][j]
r+=1
(w,ux,uy)=isotron.isotron(X,Y,30)
haty = [isotron.lin_xyz(ux,uy,dot(w,x)) for x in X]
print "Our squared error:",sum([(haty[i]-Y[i])**2 for i in range(m)])/(1.*m)
print "0's squared error:",sum([(Y[i])**2 for i in range(m)])/(1.*m)
S = zeros((n,n))
a = max(w)
b = min(w)
for i in range(n):
for j in range(n):
S[i][j]=(w[to_in(i,j)]-b)/(a-b)
sr = []
for i in range(n):
for j in range(i):
sr.append((S[i][j],chr(97+i)+chr(97+j)))
sr.sort()
res = ""
for (x,ab) in sr:
res += " "+ab
print res
return S
def compute_S(M):
n = 26
m = 2 * (26 * 26 * 27) / 2
X = zeros((m, (n * (n + 1)) / 2))
Y = zeros(m)
r = 0
for i in range(n):
for j in range(n):
for k in range(j + 1):
X[r][to_in(i, j)] = 1.
X[r][to_in(i, k)] = -1.
Y[r] = M[i][j][k]
r += 1
X[r][to_in(i, j)] = -1.
X[r][to_in(i, k)] = 1.
Y[r] = M[i][k][j]
r += 1
(w, ux, uy) = isotron.isotron(X, Y, 30)
haty = [isotron.lin_xyz(ux, uy, dot(w, x)) for x in X]
print "Our squared error:", sum([(haty[i] - Y[i])**2
for i in range(m)]) / (1. * m)
print "0's squared error:", sum([(Y[i])**2 for i in range(m)]) / (1. * m)
S = zeros((n, n))
a = max(w)
b = min(w)
for i in range(n):
for j in range(n):
S[i][j] = (w[to_in(i, j)] - b) / (a - b)
sr = []
for i in range(n):
for j in range(i):
sr.append((S[i][j], chr(97 + i) + chr(97 + j)))
sr.sort()
res = ""
for (x, ab) in sr:
res += " " + ab
print res
return S
def estimate(X, Y, X_CV, X_test, estimator, param, **kwargs):
""" Handler method to distribute estimate to correct estimation method. """
sys.path.insert(0, __path_to_source__)
sys.path.insert(0, __path_to_elm__)
assert 'estimator_id' in estimator, "estimate_handler: 'estimator_id' not in estimator"
if estimator['estimator_id'] == 'knn':
return knn(X, Y, X_CV, X_test, estimator, param, **kwargs)
elif estimator['estimator_id'] == 'linreg':
return linreg(X, Y, X_CV, X_test, estimator, param, **kwargs)
elif estimator['estimator_id'] == 'sirknn':
return sirknn(X, Y, X_CV, X_test, estimator, param, __path_to_source__,
**kwargs)
elif estimator['estimator_id'] == 'ffnn':
return ffnn(X, Y, X_CV, X_test, estimator, param, __path_to_source__,
**kwargs)
elif estimator['estimator_id'] == 'isotron':
return isotron(X, Y, X_CV, X_test, estimator, param,
__path_to_source__, **kwargs)
elif estimator['estimator_id'] == 'elm':
return elm_regressor(X, Y, X_CV, X_test, estimator, param,
__path_to_elm__, **kwargs)
else:
raise NotImplementedError("Estimator {0} is not implemented".format(
estimator.get('estimator_id', 'NONE')))
def compute_S(M):
n = 26
m = 2 * (26 * 26 * 27) / 2
X = zeros((m, (n * (n + 1)) / 2))
Y = zeros(m)
r = 0
for i in range(n):
for j in range(n):
for k in range(j + 1):
X[r][to_in(i, j)] = 1.
X[r][to_in(i, k)] = -1.
Y[r] = M[i][j][k]
r += 1
X[r][to_in(i, j)] = -1.
X[r][to_in(i, k)] = 1.
Y[r] = M[i][k][j]
r += 1
(w, ux, uy) = isotron.isotron(X, Y, 30)
haty = [isotron.lin_xyz(ux, uy, dot(w, x)) for x in X]
print "Our squared error:", sum([(haty[i] - Y[i])**2
for i in range(m)]) / (1. * m)
print "0's squared error:", sum([(Y[i])**2 for i in range(m)]) / (1. * m)
S = zeros((n, n))
a = max(w)
b = min(w)
for i in range(n):
for j in range(n):
S[i][j] = (w[to_in(i, j)] - b) / (a - b)
mms = "GraphPlot[{"
sr = []
for i in range(n):
for j in range(i):
sr.append((S[i][j], chr(97 + i) + chr(97 + j)))
if S[i][j] > 0.3:
mms += chr(97 + i) + "->" + chr(97 + j) + ","
sr.sort()
res = ""
for (x, ab) in sr:
res += " " + ab
print res
mms = mms[:-1]
mms += "},VertexLabeling->True]"
#print mms
for i in range(n):
res = chr(97 + i) + " "
sr = []
for j in range(n):
sr.append((S[i][j], chr(97 + j)))
sr.sort(reverse=True)
for (x, b) in sr:
res += " " + b
print res
# plt.plot(ux,uy)
# plt.plot(ux,[2/(1+exp(-.04*x))-1 for x in ux],'r')
# plt.show()
return S
def compute_S(M):
n = 26
m = 2*(26*26*27)/2
X = zeros((m,(n*(n+1))/2))
Y = zeros(m)
r = 0
for i in range(n):
for j in range(n):
for k in range(j+1):
X[r][to_in(i,j)]=1.
X[r][to_in(i,k)]=-1.
Y[r] = M[i][j][k]
r+=1
X[r][to_in(i,j)]=-1.
X[r][to_in(i,k)]=1.
Y[r] = M[i][k][j]
r+=1
(w,ux,uy)=isotron.isotron(X,Y,30)
haty = [isotron.lin_xyz(ux,uy,dot(w,x)) for x in X]
print "Our squared error:",sum([(haty[i]-Y[i])**2 for i in range(m)])/(1.*m)
print "0's squared error:",sum([(Y[i])**2 for i in range(m)])/(1.*m)
S = zeros((n,n))
a = max(w)
b = min(w)
for i in range(n):
for j in range(n):
S[i][j]=(w[to_in(i,j)]-b)/(a-b)
mms = "GraphPlot[{"
sr = []
for i in range(n):
for j in range(i):
sr.append((S[i][j],chr(97+i)+chr(97+j)))
if S[i][j]>0.3:
mms+=chr(97+i)+"->"+chr(97+j)+","
sr.sort()
res = ""
for (x,ab) in sr:
res += " "+ab
print res
mms=mms[:-1]
mms+="},VertexLabeling->True]"
#print mms
for i in range(n):
res = chr(97+i)+" "
sr = []
for j in range(n):
sr.append((S[i][j],chr(97+j)))
sr.sort(reverse=True)
for (x,b) in sr:
res += " "+b
print res
# plt.plot(ux,uy)
# plt.plot(ux,[2/(1+exp(-.04*x))-1 for x in ux],'r')
# plt.show()
return S
