本文整理匯總了Python中mlp.MLP.mean_square_error方法的典型用法代碼示例。如果您正苦於以下問題：Python MLP.mean_square_error方法的具體用法？Python MLP.mean_square_error怎麽用？Python MLP.mean_square_error使用的例子？那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在類mlp.MLP的用法示例。
在下文中一共展示了MLP.mean_square_error方法的1個代碼示例，這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚，您的評價將有助於係統推薦出更棒的Python代碼示例。
示例1: test_cA
# 需要導入模塊: from mlp import MLP [as 別名]
# 或者: from mlp.MLP import mean_square_error [as 別名]
def test_cA(learning_rate=0.002, training_epochs=10,
dataset='test2.pkl.gz', n_epochs=100,
batch_size=5, contraction_level=0.001):
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] / batch_size
# validation/testの時はミニバッチを使わない
n_valid_batches = valid_set_x.get_value(borrow=True).shape[0] #/ batch_size
n_test_batches = test_set_x.get_value(borrow=True).shape[0] #/ batch_size
print '... building the model'
# 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.vector('y') # the labels are presented as 1D vector of
# [int] labels
# 亂數シード
rng = np.random.RandomState(1234)
regressor = MLP(rng=rng, input=x, n_in=1631,
n_hidden=800)
ca = cA(numpy_rng=rng, input=x,
n_visible=1631, n_hidden=800, n_batchsize=batch_size,
W = regressor.hiddenLayer.W, bhid = regressor.hiddenLayer.b)
cost, updates = ca.get_cost_updates(contraction_level=contraction_level,
learning_rate=0.001)
train_ca = theano.function([index], [T.mean(ca.L_rec), ca.L_jacob],
updates=updates,
givens={x: train_set_x[index * batch_size:
(index + 1) * batch_size]})
start_time = time.clock()
############
# TRAINING #
############
print '... training the model'
epoch = 0
# go through training epochs
for epoch in xrange(training_epochs):
# go through trainng set
c = []
for batch_index in xrange(n_train_batches):
c.append(train_ca(batch_index))
c_array = np.vstack(c)
print 'Training epoch %d, reconstruction cost ' % epoch, np.mean(
c_array[0]), ' jacobian norm ', np.mean(np.sqrt(c_array[1]))
a = regressor.hiddenLayer.W.eval()
print np.sum(a)
end_time = time.clock()
training_time = (end_time - start_time)
print >> sys.stderr, ('The code for file ' + os.path.split(__file__)[1] +
' ran for %.2fm' % ((training_time) / 60.))
cost = regressor.mean_square_error(y)
gparams = []
for param in regressor.params:
gparam = T.grad(cost, param)
gparams.append(gparam)
updates = []
for param, gparam in zip(regressor.params, gparams):
updates.append((param, param - learning_rate * gparam))
# 學習
train_model = theano.function(inputs=[index], outputs=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]})
# validation
validate_model = theano.function(inputs=[index],
outputs=regressor.prediction,
givens={
x: valid_set_x[index:(index + 1)]})
# validationデータセット
valid_datasets = load_valid_data(dataset)
#.........這裏部分代碼省略.........