def __init__(self, x, y, batch_size, videos, kernels, pools, n_input, n_output, hidden_input, params=None):
learning_rate = 0.1
rng = numpy.random.RandomState(1234)
print '... building the model'
sys.stdout.flush()
if not params:
# Construct the first convolutional pooling layer:
# filtering reduces the image size to (28-5+1,28-5+1)=(24,24)
# maxpooling reduces this further to (24/2,24/2) = (12,12)
# 4D output tensor is thus of shape (batch_size,nkerns[0],12,12)
layer0 = ConvLayer(x, n_input[0], n_output[0], kernels[0], videos[0], pools[0],
batch_size, 'L0', rng)
layer1 = ConvLayer(layer0.output, n_input[1], n_output[1], kernels[1], videos[1], pools[1],
batch_size, 'L1', rng)
layer2_input = layer1.output.flatten(2)
# construct a fully-connected sigmoidal layer
layer2 = HiddenLayer(rng, input=layer2_input, n_in=hidden_input,
n_out=batch_size, activation=T.tanh)
# classify the values of the fully-connected sigmoidal layer
layer3 = LogisticRegression(input=layer2.output, n_in=batch_size, n_out=2)
else:
layer0 = ConvLayer(x, n_input[0], n_output[0], kernels[0], videos[0], pools[0],
batch_size, 'L0', rng, True, params[6], params[7])
layer1 = ConvLayer(layer0.output, n_input[1], n_output[1], kernels[1], videos[1], pools[1],
batch_size, 'L1', rng, True, params[4], params[5])
layer2_input = layer1.output.flatten(2)
# construct a fully-connected sigmoidal layer
layer2 = HiddenLayer(rng, input=layer2_input, n_in=hidden_input,
n_out=batch_size, activation=T.tanh, W=params[2], b=params[3])
# classify the values of the fully-connected sigmoidal layer
layer3 = LogisticRegression(input=layer2.output, n_in=batch_size, n_out=2, W=params[0], b=params[1])
# the cost we minimize during training is the NLL of the model
cost = layer3.negative_log_likelihood(y)
# create a list of all model parameters to be fit by gradient descent
self.params = layer3.params + layer2.params + layer1.params + layer0.params
# create a list of gradients for all model parameters
grads = T.grad(cost, self.params)
# train_model is a function that updates the model parameters by
# SGD Since this model has many parameters, it would be tedious to
# manually create an update rule for each model parameter. We thus
# create the updates list by automatically looping over all
# (params[i],grads[i]) pairs.
updates = []
for param_i, grad_i in zip(self.params, grads):
updates.append((param_i, param_i - learning_rate * grad_i))
self.train_model = theano.function([x, y], cost, updates=updates)
self.validate_model = theano.function(inputs=[x, y], outputs=layer3.errors(y))
self.predict = theano.function(inputs=[x], outputs=layer3.y_pred)
print '... building done'
sys.stdout.flush()
def evaluate_lenet5(learning_rate=0.1, n_epochs=200,
dataset='mnist.pkl.gz',
nkerns=[20, 50], batch_size=500):
""" Demonstrates lenet on MNIST dataset
:type learning_rate: float
:param learning_rate: learning rate used (factor for the stochastic
gradient)
:type n_epochs: int
:param n_epochs: maximal number of epochs to run the optimizer
:type dataset: string
:param dataset: path to the dataset used for training /testing (MNIST here)
:type nkerns: list of ints
:param nkerns: number of kernels on each layer
"""
rng = numpy.random.RandomState(23455)
datasets = load_data(dataset)
train_set_x, train_set_y = datasets[0]
valid_set_x, valid_set_y = datasets[1]
test_set_x, test_set_y = datasets[2]
# compute number of minibatches for training, validation and testing
n_train_batches = train_set_x.get_value(borrow=True).shape[0]
n_valid_batches = valid_set_x.get_value(borrow=True).shape[0]
n_test_batches = test_set_x.get_value(borrow=True).shape[0]
n_train_batches /= batch_size
n_valid_batches /= batch_size
n_test_batches /= batch_size
# allocate symbolic variables for the data
index = T.lscalar() # index to a [mini]batch
# start-snippet-1
x = T.matrix('x') # the data is presented as rasterized images
y = T.ivector('y') # the labels are presented as 1D vector of
# [int] labels
######################
# BUILD ACTUAL MODEL #
######################
print '... building the model'
# Reshape matrix of rasterized images of shape (batch_size, 28 * 28)
# to a 4D tensor, compatible with our LeNetConvPoolLayer
# (28, 28) is the size of MNIST images.
layer0_input = x.reshape((batch_size, 1, 28, 28))
# Construct the first convolutional pooling layer:
# filtering reduces the image size to (28-5+1 , 28-5+1) = (24, 24)
# maxpooling reduces this further to (24/2, 24/2) = (12, 12)
# 4D output tensor is thus of shape (batch_size, nkerns[0], 12, 12)
layer0 = LeNetConvPoolLayer(
rng,
input=layer0_input,
image_shape=(batch_size, 1, 28, 28),
filter_shape=(nkerns[0], 1, 5, 5),
poolsize=(2, 2)
)
# Construct the second convolutional pooling layer
# filtering reduces the image size to (12-5+1, 12-5+1) = (8, 8)
# maxpooling reduces this further to (8/2, 8/2) = (4, 4)
# 4D output tensor is thus of shape (batch_size, nkerns[1], 4, 4)
layer1 = LeNetConvPoolLayer(
rng,
input=layer0.output,
image_shape=(batch_size, nkerns[0], 12, 12),
filter_shape=(nkerns[1], nkerns[0], 5, 5),
poolsize=(2, 2)
)
# the HiddenLayer being fully-connected, it operates on 2D matrices of
# shape (batch_size, num_pixels) (i.e matrix of rasterized images).
# This will generate a matrix of shape (batch_size, nkerns[1] * 4 * 4),
# or (500, 50 * 4 * 4) = (500, 800) with the default values.
layer2_input = layer1.output.flatten(2)
# construct a fully-connected sigmoidal layer
layer2 = HiddenLayer(
rng,
input=layer2_input,
n_in=nkerns[1] * 4 * 4,
n_out=500,
activation=T.tanh
)
# classify the values of the fully-connected sigmoidal layer
layer3 = LogisticRegression(input=layer2.output, n_in=500, n_out=10)
# the cost we minimize during training is the NLL of the model
cost = layer3.negative_log_likelihood(y)
# create a function to compute the mistakes that are made by the model
test_model = theano.function(
[index],
layer3.errors(y),
givens={
x: test_set_x[index * batch_size: (index + 1) * batch_size],
y: test_set_y[index * batch_size: (index + 1) * batch_size]
}
)
validate_model = theano.function(
[index],
layer3.errors(y),
givens={
x: valid_set_x[index * batch_size: (index + 1) * batch_size],
y: valid_set_y[index * batch_size: (index + 1) * batch_size]
}
)
# create a list of all model parameters to be fit by gradient descent
params = layer3.params + layer2.params + layer1.params + layer0.params
# create a list of gradients for all model parameters
grads = T.grad(cost, params)
# train_model is a function that updates the model parameters by
# SGD Since this model has many parameters, it would be tedious to
# manually create an update rule for each model parameter. We thus
# create the updates list by automatically looping over all
# (params[i], grads[i]) pairs.
updates = [
(param_i, param_i - learning_rate * grad_i)
for param_i, grad_i in zip(params, grads)
]
train_model = theano.function(
[index],
cost,
updates=updates,
givens={
x: train_set_x[index * batch_size: (index + 1) * batch_size],
y: train_set_y[index * batch_size: (index + 1) * batch_size]
}
)
# end-snippet-1
###############
# TRAIN MODEL #
###############
print '... training'
# early-stopping parameters
patience = 10000 # look as this many examples regardless
patience_increase = 2 # wait this much longer when a new best is
# found
improvement_threshold = 0.995 # a relative improvement of this much is
# considered significant
validation_frequency = min(n_train_batches, patience / 2)
# go through this many
# minibatche before checking the network
# on the validation set; in this case we
# check every epoch
best_validation_loss = numpy.inf
best_iter = 0
test_score = 0.
start_time = timeit.default_timer()
epoch = 0
done_looping = False
while (epoch < n_epochs) and (not done_looping):
epoch = epoch + 1
for minibatch_index in xrange(n_train_batches):
iter = (epoch - 1) * n_train_batches + minibatch_index
if iter % 100 == 0:
print 'training @ iter = ', iter
cost_ij = train_model(minibatch_index)
if (iter + 1) % validation_frequency == 0:
# compute zero-one loss on validation set
validation_losses = [validate_model(i) for i
in xrange(n_valid_batches)]
this_validation_loss = numpy.mean(validation_losses)
print('epoch %i, minibatch %i/%i, validation error %f %%' %
(epoch, minibatch_index + 1, n_train_batches,
this_validation_loss * 100.))
# if we got the best validation score until now
if this_validation_loss < best_validation_loss:
#improve patience if loss improvement is good enough
if this_validation_loss < best_validation_loss * \
improvement_threshold:
patience = max(patience, iter * patience_increase)
# save best validation score and iteration number
best_validation_loss = this_validation_loss
best_iter = iter
# test it on the test set
test_losses = [
test_model(i)
for i in xrange(n_test_batches)
]
test_score = numpy.mean(test_losses)
print((' epoch %i, minibatch %i/%i, test error of '
'best model %f %%') %
(epoch, minibatch_index + 1, n_train_batches,
test_score * 100.))
if patience <= iter:
done_looping = True
break
end_time = timeit.default_timer()
print('Optimization complete.')
print('Best validation score of %f %% obtained at iteration %i, '
'with test performance %f %%' %
(best_validation_loss * 100., best_iter + 1, test_score * 100.))
print >> sys.stderr, ('The code for file ' +
os.path.split(__file__)[1] +
' ran for %.2fm' % ((end_time - start_time) / 60.))
def train_lenet5(train_set_x, train_set_y, params, batch_size,
learning_rate=0.01, nkerns=[20,50], test=False):
"""
Trains LeNet-5 on MNIST dataset, and returns trained parameters on
completion.
:type train_set: list of floats
:param train_set: training samples (x- and y-values) for training
:type learning_rate: float
:param learning_rate: learning rate used (factor for the stochastic
gradient)
:type params: list of tuples of floats
:param params: list of tuple of parameters from Supervisor. Takes the form
[(W_layer0, b_layer0), (W_layer1, b_layer1),
(W_layer2, b_layer2), (W_layer3, b_layer3)]
:type batch_size: int
:param batch_size: size of training batch
:type nkerns: list of ints
:param nkerns: number of kernels on each layer
Output: tuple of LeNet-5 parameters by layer, in this format:
( (W_layer0, b_layer0), ..., (W_layer3, b_layer3) )
"""
rng = numpy.random.RandomState(23455)
# compute number of minibatches for training, validation and testing
n_train_batches = 100
#n_train_batches = train_set_x.get_value(borrow=True).shape[0] / batch_size
# allocate symbolic variables for the data
index = T.lscalar() # index to a [mini]batch
x = T.matrix('x') # the data is presented as rasterized images
y = T.ivector('y') # the labels are presented as 1D vector of
# [int] labels
ishape = (28,28) # this is the size of MNIST images
print ' ... building the model'
# Reshape matrix of rasterized images of shape (batch_size,28*28)
# to a 4D tensor, compatible with our LeNetConvPoolLayer
layer0_input = x.reshape((batch_size,1,28,28))
# Construct the first convolutional pooling layer:
# filtering reduces the image size to (28-5+1,28-5+1)=(24,24)
# maxpooling reduces this further to (24/2,24/2) = (12,12)
# 4D output tensor is thus of shape (batch_size,nkerns[0],12,12)
layer0 = LeNetConvPoolLayer(rng, input=layer0_input,
image_shape=(batch_size,1,28,28),
W_values = params[0][0],
b_values = params[0][1],
filter_shape=(nkerns[0],1,5,5), poolsize=(2,2))
# Construct the second convolutional pooling layer
# filtering reduces the image size to (12-5+1,12-5+1)=(8,8)
# maxpooling reduces this further to (8/2,8/2) = (4,4)
# 4D output tensor is thus of shape (nkerns[0],nkerns[1],4,4)
layer1 = LeNetConvPoolLayer(rng, input=layer0.output,
image_shape=(batch_size,nkerns[0],12,12),
W_values = params[1][0],
b_values = params[1][1],
filter_shape=(nkerns[1],nkerns[0],5,5),
poolsize=(2,2))
# the TanhLayer being fully-connected, it operates on 2D matrices of
# shape (batch_size,num_pixels) (i.e matrix of rasterized images).
# This will generate a matrix of shape (20,32*4*4) = (20,512)
layer2_input = layer1.output.flatten(2)
# construct a fully-connected sigmoidal layer
layer2 = HiddenLayer(rng, input=layer2_input, n_in=nkerns[1]*4*4,
n_out=120, activation = T.tanh,
W_values = params[2][0],
b_values = params[2][1])
# classify the values of the fully-connected sigmoidal layer
layer3 = LogisticRegression(input=layer2.output, n_in=120, n_out=10,
W_values=params[3][0],
b_values=params[3][1])
# the cost we minimize during training is the NLL of the model
cost = layer3.negative_log_likelihood(y)
# create a list of all model parameters to be fit by gradient descent
params = layer3.params + layer2.params + layer1.params + layer0.params
# create a list of gradients for all model parameters
grads = T.grad(cost, params)
# train_model is a function that updates the model parameters by SGD
# Since this model has many parameters, it would be tedious to manually
# create an update rule for each model parameter. We thus create the updates
# dictionary by automatically looping over all (params[i],grads[i]) pairs.
updates = {}
for param_i, grad_i in zip(params, grads):
updates[param_i] = param_i - learning_rate * grad_i
train_model = theano.function([index], cost, updates=updates,
givens = {x: train_set_x, y: train_set_y},
mode='FAST_RUN')
print " training lenet-5..."
start_time = time.clock()
epoch = 0
done_looping = False
while (epoch < batch_size/100) and (not done_looping):
epoch = epoch + 1
for minibatch_index in xrange(n_train_batches):
iter = epoch * n_train_batches + minibatch_index
cost_ij = train_model(minibatch_index)
end_time = time.clock()
print " worker training complete."
print " %i samples analyzed in %.2fm" % (batch_size,
(end_time-start_time)/60.)
return ((layer0.params[0].get_value(), layer0.params[1].get_value()),
(layer1.params[0].get_value(), layer1.params[1].get_value()),
(layer2.params[0].get_value(), layer2.params[1].get_value()),
(layer3.params[0].get_value(), layer3.params[1].get_value()))
