def pretraining_functions(self, train_set_x, batch_size):
# index to a [mini]batch
index = T.lscalar('index') # index to a minibatch
corruption_level = T.scalar('corruption') # % of corruption to use
learning_rate = T.scalar('lr') # learning rate to use
momentum = T.scalar('momentum')
batch_size = batch_size if train_set_x.get_value(borrow=True).shape[0] > batch_size else train_set_x.get_value(borrow=True).shape[0]
# number of batches
n_batches = train_set_x.get_value(borrow=True).shape[0] / batch_size
# begining of a batch, given `index`
batch_begin = index * batch_size
# ending of a batch given `index`
batch_end = batch_begin + batch_size
pretrain_fns = []
for dA in self.dA_layers:
# get the cost and the updates list
cost, updates = dA.get_cost_updates(corruption_level, learning_rate, momentum)
# compile the theano function
fn = theano.function(inputs=[index,
theano.Param(corruption_level, default=0.2),
theano.Param(learning_rate, default=0.1)
,theano.Param(momentum, default=0.5)
],
outputs=cost,
updates=updates,
givens={self.x: train_set_x[batch_begin:batch_end]})
# append `fn` to the list of functions
pretrain_fns.append(fn)
return pretrain_fns
def pretraining_function(self, dA, train_set_x, batch_size):
# index to a [mini]batch
index = T.lscalar('index') # index to a minibatch
corruption_level = T.scalar('corruption') # % of corruption to use
learning_rate = T.scalar('lr') # learning rate to use
momentum = T.scalar('momentum')
# number of batches
batch_size = batch_size if train_set_x.get_value(borrow=True).shape[0] > batch_size else train_set_x.get_value(borrow=True).shape[0]
n_batches = train_set_x.get_value(borrow=True).shape[0] / batch_size
# begining of a batch, given `index`
batch_begin = index * batch_size
# ending of a batch given `index`
batch_end = batch_begin + batch_size
cost, updates = dA.get_cost_updates(corruption_level, learning_rate, momentum, weight_y = self.cfg.settings['weight_y'] , size_of_x = self.cfg.n_ins, size_of_y = self.cfg.n_outs)
# compile the theano function
fn = theano.function(inputs=[index,
theano.Param(corruption_level, default=0.2),
theano.Param(learning_rate, default=0.1)
,theano.Param(momentum, default=0.5)
],
outputs=cost,
updates=updates,
givens={dA.x: train_set_x[batch_begin:batch_end]})
return fn
def get_cost_functions(self, train_set_x):
# get cost of train_set_x or validation set for each dA layer.
cost_fns = []
for dA in self.dA_layers:
# get the cost and the updates list
cost, updates = dA.get_cost_updates(0, 0)
# compile the theano function
fn = theano.function(inputs=[],
outputs=cost,
givens={self.x: train_set_x})
# append `fn` to the list of functions
cost_fns.append(fn)
