def backprop(self, x, y):
nabla_b = [np.zeros(b.shape) for b in self.biases]
nabla_w = [np.zeros(w.shape) for w in self.weights]
activation = x
activations = [x]
zs = []
for b, w in zip(self.biases, self.weights):
#print w
#print len(activation)
#print len(w)
z = np.dot(w, activation) + b
#print z
zs.append(z)
activation = myalgorithm.sigmoid(z)
#print activation
activations.append(activation)
#print activations[-1]
#print y
delta = self.cost_derivative(activations[-1],
y) * myalgorithm.sigmoid_prime(zs[-1])
nabla_b[-1] = delta
nabla_w[-1] = np.dot(delta, activations[-2].transpose())
for l in xrange(2, self.num_layers):
z = zs[-l]
sp = myalgorithm.sigmoid_prime(z)
delta = np.dot(self.weights[-l + 1].transpose(), delta) * sp
nabla_b[-l] = delta
nabla_w[-l] = np.dot(delta, activations[-l - 1].transpose())
return (nabla_b, nabla_w)
def feedforward(self, a):
for b, w in zip(self.biases, self.weights):
a = myalgorithm.sigmoid(np.dot(w, a) + b)
# print a
return a
def outcome(self, a):
for b, w in zip(self.biases, self.weights):
a = myalgorithm.sigmoid(np.dot(w, a) + b)
return np.argmax(a)
def outcome(weights, biases, a):
for b, w in zip(biases, weights):
a = myalgorithm.sigmoid(numpy.dot(w, a) + b)
return numpy.argmax(a)