def __init__(self, layers_shape, bias_unit=True):
""" initialisation of the ff neural network"""
# check minimum layer shape
if len(layers_shape) < 2:
raise FeedForwardNeuralNetworkError, "can't init a neural net without at least the in and out size"
# save the layer shape
self._layers_shape = layers_shape
self._layers_count = len(layers_shape)
self._bias_unit = bias_unit
# init layer input/output memory
self._layer_input = list()
self._layer_output = list()
# init the neural layer
self._weights = []
for s_in, s_out in zip(layers_shape[:-1], layers_shape[1:]):
# random init with mean 0 of a layer (with bias unit if require)
self._weights.append(2 * Matrix.random(s_in + bias_unit, s_out) - 1)
from MatrixVector import Matrix, Vector
##### 2 Layer Neural Net
# input data
X = Matrix([Vector([0.,0.,1.]),
Vector([0.,1.,1.]),
Vector([1.,0.,1.]),
Vector([1.,1.,1.])])
y = Matrix([Vector([0.,1.,1.,0.])]).transpose()
# randomly initialize our weights with mean 0
syn0 = 2 * Matrix.random(3, 4) - 1
syn1 = 2 * Matrix.random(4, 1) - 1
for j in xrange(60000):
# Feed forward through layers 0, 1, and 2
l0 = X
l1 = l0.dotProduct(syn0).nonlin()
l2 = l1.dotProduct(syn1).nonlin()
# how much did we miss the target value?
l2_error = y - l2
if (j% 10000) == 0:
print "Error:" + str(l2_error.abs().mean())
# in what direction is the target value?
from MatrixVector import Matrix, Vector
##### 2 Layer Neural Net
# input data
X = Matrix([Vector([0.,0.,1.]),
Vector([0.,1.,1.]),
Vector([1.,0.,1.]),
Vector([1.,1.,1.])])
y = Matrix([Vector([0.,1.,1.,0.]), Vector([0.,1.,0.,0.])]).transpose()
# randomly initialize our weights with mean 0
syn0 = 2 * Matrix.random(3, 4) - 1
syn1 = 2 * Matrix.random(4, 2) - 1
for j in xrange(60000):
# Feed forward through layers 0, 1, and 2
l0 = X
l1 = l0.dotProduct(syn0).nonlin()
l2 = l1.dotProduct(syn1).nonlin()
# how much did we miss the target value?
l2_error = y - l2
if (j% 10000) == 0:
print "Error:" + str(l2_error.abs().mean())
# in what direction is the target value?
from MatrixVector import Matrix, Vector
##### 2 Layer Neural Net
# input data
X = Matrix([Vector([0.0, 0.0, 1.0]), Vector([0.0, 1.0, 1.0]), Vector([1.0, 0.0, 1.0]), Vector([1.0, 1.0, 1.0])])
Xones = Matrix(Matrix.ones(1, X.getRowLen())._matrix + X.transpose()._matrix).transpose()
y = Matrix([Vector([0.0, 1.0, 1.0, 0.0])]).transpose()
# randomly initialize our weights with mean 0 (entry + 1 for bias unit)
syn0 = 2 * Matrix.random(4, 4) - 1
syn1 = 2 * Matrix.random(5, 1) - 1
# set the alpha
alpha = 10
for j in xrange(1000):
# Feed forward through layers 0, 1, and 2
l0 = Xones
l1 = l0.dotProduct(syn0).nonlin()
l1 = Matrix(Matrix.ones(1, l0.getRowLen())._matrix + l1.transpose()._matrix).transpose()
l2 = l1.dotProduct(syn1).nonlin()
# how much did we miss the target value?
l2_error = y - l2
if (j % 100) == 0:
print "Error:" + str(l2_error.abs().mean())
from MatrixVector import Matrix, Vector
##### 2 Layer Neural Net
# input data
X = Matrix([Vector([0.,0.,1.]),
Vector([0.,1.,1.]),
Vector([1.,0.,1.]),
Vector([1.,1.,1.])])
y = Matrix([Vector([0.,0.,1.,1.])]).transpose()
syn0 = 2 * Matrix.random(3,1) - 1
print syn0
for iter in xrange(10000):
# forward propagation
l0 = X
l1 = (l0.dotProduct(syn0)).nonlin()
# error eval
l1_error = y - l1
# multiply how much we missed by the
# slope of the sigmoid at the values in l1
l1_delta = l1_error * l1.nonlin(True)
# weight update
syn0 = syn0 + (l0.transpose()).dotProduct(l1_delta)
def append_bias(mtx):
return Matrix(Matrix.ones(1, mtx.getRowLen())._matrix + mtx.transpose()._matrix).transpose()
from MatrixVector import Matrix, Vector
def append_bias(mtx):
return Matrix(Matrix.ones(1, mtx.getRowLen())._matrix + mtx.transpose()._matrix).transpose()
def remove_bias(mtx):
return Matrix(mtx.transpose()._matrix[1:]).transpose()
##### 4 Layer Neural Net
# input data
X = Matrix([Vector([0.0, 0.0, 1.0]), Vector([0.0, 1.0, 1.0]), Vector([1.0, 0.0, 1.0]), Vector([1.0, 1.0, 1.0])])
Xones = Matrix(Matrix.ones(1, X.getRowLen())._matrix + X.transpose()._matrix).transpose()
y = Matrix([Vector([0.0, 1.0, 1.0, 0.0])]).transpose()
# randomly initialize our weights with mean 0 (entry + 1 for bias unit)
syn0 = 2 * Matrix.random(4, 4) - 1
syn1 = 2 * Matrix.random(5, 4) - 1
syn2 = 2 * Matrix.random(5, 1) - 1
# set the alpha
alpha = 0.07
for j in xrange(10000):
# Feed forward through layers 0, 1, 2 and 3