def one_vs_all(_X, y, num_labels, lambda_):
m, n = _X.shape
X = np.hstack((np.ones((m, 1)), _X))
all_theta = []
for c in range(num_labels):
if c == 0: c = 10
def costf(theta):
return logistic_cost_function(theta, X, y == c, lambda_)
def gradf(theta):
return logistic_grad_function(theta, X, y == c, lambda_)
maxiter = 50
initial_theta = np.zeros(n + 1)
theta, allvec = opt.fmin_ncg(costf,
initial_theta,
gradf,
maxiter=maxiter,
retall=1,
callback=step())
# Jhist = map(costf, allvec)
# pl.plot(Jhist)
# pl.show()
all_theta.append(theta)
return np.vstack(all_theta)
def on_mouse_motion(self, start, end, buttons, start_hovered, end_hovered):
if buttons[0] == buttons[2]:
return
row = int((end[1] - self._bw) // (self._ts + self._lw))
col = int((end[0] - self._bw) // (self._ts + self._lw))
if not (0 <= row < self.nrows and 0 <= col < self.ncols) or (row, col) == self._previous:
return
current = row, col
if buttons[0]:
if self.color != self.at(current).color \
or self.mark != self.at(current).mark:
self.erase(current)
self._visited.discard(current)
if self.connection_mode == Grid.TREE:
if current not in self._visited:
adj = util.adjacency(current, self._previous)
self.connect(self._previous, adj)
elif self.connection_mode == Grid.BLOB:
for direction in NORTH, WEST, SOUTH, EAST:
if util.step(current, direction) in self._visited:
self.connect(current, direction)
elif self.connection_mode == Grid.TRACE:
self.connect(self._previous, util.adjacency(current, self._previous))
self.put(current, self.color, self.mark)
self._visited.add(current)
else:
self.erase(current)
self._visited.discard(current)
self._previous = current
def flood_set(grid, p):
queue = [p]
visited = set()
while queue:
q = queue.pop()
visited.add(q)
for direction in (NORTH, WEST, SOUTH, EAST):
adj = util.step(q, direction)
if grid.at(q).connections & direction and adj not in visited:
queue.append(adj)
return visited
def ex2_reg():
data = load_txt(os.path.join(ex2path, 'ex2data2.txt'))
_X = data[:, :2]
y = data[:, 2]
# plot_data(_X, y)
# pl.xlabel('Microchip Test 1')
# pl.ylabel('Microchip Test 2')
# pl.legend(['y=1', 'y=0'])
X = map_feature(_X[:, 0], _X[:, 1])
initial_theta = np.zeros(X.shape[1])
lambda_ = 0
print 'cost at initial theta (zeros):', cost_function(
initial_theta, X, y, lambda_)[0]
# Regularization
#X_norm, mu, sigma = feature_normalize(X[:,1:])
#X = np.hstack((np.ones((m, 1)), X_norm))
def costf(theta):
return logistic_cost_function(theta, X, y, lambda_)
def difff(theta):
return logistic_grad_function(theta, X, y, lambda_)
maxiter = 50
theta, allvec = opt.fmin_ncg(costf,
initial_theta,
difff,
retall=1,
maxiter=maxiter,
callback=step())
# theta, allvec = opt.fmin_bfgs(costf, initial_theta, difff, retall=1, maxiter=maxiter, callback=step())
print 'optimal cost:', costf(theta)
Jhist = [costf(t) for t in allvec]
pl.figure()
pl.plot(Jhist)
plot_decision_boundary(theta, X, y)
# Compute accuracy on our training set
h = np.dot(X, theta)
print 'Train Accuracy:', ((h > 0) == y).mean() * 100
def one_vs_all(_X, y, num_labels, lambda_):
m, n = _X.shape
X = np.hstack((np.ones((m, 1)), _X))
all_theta = []
for c in range(num_labels):
if c == 0: c = 10
def costf(theta):
return logistic_cost_function(theta, X, y==c, lambda_)
def gradf(theta):
return logistic_grad_function(theta, X, y==c, lambda_)
maxiter = 50
initial_theta = np.zeros(n+1)
theta, allvec = opt.fmin_ncg(costf, initial_theta, gradf, maxiter=maxiter, retall=1, callback=step())
# Jhist = map(costf, allvec)
# pl.plot(Jhist)
# pl.show()
all_theta.append(theta)
return np.vstack(all_theta)