def __init__(self,numin,numclasses):
self.numin = numin
self.numclasses = numclasses
self.params = 0.01 * randn(self.numin*self.numclasses+self.numclasses)
self.scorefunc = logreg_score(self.numin,self.numclasses,self.params)
self.scorefuncs = [scorefunc]
Contrastive.__init__(self,normalizeacrosscliques=False)
def __init__(self,numin,numout):
self.numin = numin
self.numout = numout
self.params = 0.01 * randn(numin*numout+numout)
self.scorefunc = scorefunc.LinearRegressionScore(numin,numout,self.params)
self.scorefuncs = [self.scorefunc]
Contrastive.__init__(self,normalizeacrosscliques=False)
def __init__(self, numin, numclasses):
self.numin = numin
self.numclasses = numclasses
self.params = 0.01 * randn(self.numin * self.numclasses +
self.numclasses)
self.scorefunc = logreg_score(self.numin, self.numclasses, self.params)
self.scorefuncs = [scorefunc]
Contrastive.__init__(self, normalizeacrosscliques=False)
def __init__(self, numin, numout):
self.numin = numin
self.numout = numout
self.params = 0.01 * randn(numin * numout + numout)
self.scorefunc = scorefunc.LinearRegressionScore(
numin, numout, self.params)
self.scorefuncs = [self.scorefunc]
Contrastive.__init__(self, normalizeacrosscliques=False)
def __init__(self,numin,numhid1,numhid2,numhid3,numout):
self.numin = numin
self.numhid1 = numhid1
self.numhid2 = numhid2
self.numhid3 = numhid3
self.numout = numout
self.params = 0.1 * randn(scorefunc.Islsl.numparams(\
numin,numhid1,numhid2,numhid3,numout))
self.scorefuncs = [scorefunc.Islsl(\
numin,numhid1,numhid2,numhid3,numout,self.params)]
Contrastive.__init__(self,normalizeacrosscliques=False)
def __init__(self,numin,numhid,numout,params=None):
self.numin = numin
self.numhid = numhid
self.numout = numout
self.params = params
if self.params == None:
self.params = 0.01 * randn(self.numin*self.numhid+self.numhid+\
self.numhid*self.numout+self.numout)
self.scorefuncs = [scorefunc.SigmoidhiddenLinearoutputScore\
(numin,numhid,numout,self.params)]
Contrastive.__init__(self,normalizeacrosscliques=False)
def __init__(self, numin, numhid1, numhid2, numhid3, numout):
self.numin = numin
self.numhid1 = numhid1
self.numhid2 = numhid2
self.numhid3 = numhid3
self.numout = numout
self.params = 0.1 * randn(scorefunc.Islsl.numparams(\
numin,numhid1,numhid2,numhid3,numout))
self.scorefuncs = [scorefunc.Islsl(\
numin,numhid1,numhid2,numhid3,numout,self.params)]
Contrastive.__init__(self, normalizeacrosscliques=False)
def __init__(self, numin, numhid, numout, params=None):
self.numin = numin
self.numhid = numhid
self.numout = numout
self.params = params
if self.params == None:
self.params = 0.01 * randn(self.numin*self.numhid+self.numhid+\
self.numhid*self.numout+self.numout)
self.scorefuncs = [scorefunc.SigmoidhiddenLinearoutputScore\
(numin,numhid,numout,self.params)]
Contrastive.__init__(self, normalizeacrosscliques=False)
def __init__(self, n_nodes):
self.n_nodes = zip(n_nodes[:-1], n_nodes[1:])
self.params = 0.01 * randn(sum(map(lambda (a,b): (a+1)*b + (b+1)*a, self.n_nodes)))
self.layers = []
start, stop = 0, self.params.shape[0]
for (numin, numhid) in self.n_nodes:
_in = (numin+1)*numhid
_out = (numhid+1)*numin
assert start+_in <= stop-_out, (start, _in, stop, _out)
idx = range(start, start+_in) + range(stop-_out, stop)
start += _in
stop -= _out
self.layers += [Isl(numin, numhid, numin, self.params[idx])]
Contrastive.__init__(self, normalizeacrosscliques=False)
def __init__(self,numvis,numhid,sparsitygain=0.0,targethidprobs=0.2,\
cditerations=1,normalizeacrosscliques=True,\
meanfield_output=True,verbose=False):
self.targethidprobs = targethidprobs
self.numvis = numvis
self.numhid = numhid
self.cditerations = cditerations
self.sparsitygain = sparsitygain
self.meanfield_output = meanfield_output
self.params = 0.01*randn(numvis*numhid+numvis+numhid)
self.wyh = asmatrix(reshape(self.params[:numvis*numhid],(numvis,numhid)))
self.wy = asmatrix(self.params[numvis*numhid:numvis*numhid+numvis]).T
self.wh = asmatrix(self.params[numvis*numhid+numvis:]).T
self.verbose = verbose
Contrastive.__init__(self,normalizeacrosscliques)
def __init__(self, n_nodes):
self.n_nodes = zip(n_nodes[:-1], n_nodes[1:])
self.params = 0.01 * randn(
sum(map(lambda (a, b): (a + 1) * b + (b + 1) * a, self.n_nodes)))
self.layers = []
start, stop = 0, self.params.shape[0]
for (numin, numhid) in self.n_nodes:
_in = (numin + 1) * numhid
_out = (numhid + 1) * numin
assert start + _in <= stop - _out, (start, _in, stop, _out)
idx = range(start, start + _in) + range(stop - _out, stop)
start += _in
stop -= _out
self.layers += [Isl(numin, numhid, numin, self.params[idx])]
Contrastive.__init__(self, normalizeacrosscliques=False)
def __init__(self,numvis,numhid,sparsitygain=0.0,targethidprobs=0.2,\
cditerations=1,normalizeacrosscliques=True,\
meanfield_output=True,verbose=False):
self.targethidprobs = targethidprobs
self.numvis = numvis
self.numhid = numhid
self.cditerations = cditerations
self.sparsitygain = sparsitygain
self.meanfield_output = meanfield_output
self.params = 0.01 * randn(numvis * numhid + numvis + numhid)
self.wyh = asmatrix(
reshape(self.params[:numvis * numhid], (numvis, numhid)))
self.wy = asmatrix(self.params[numvis * numhid:numvis * numhid +
numvis]).T
self.wh = asmatrix(self.params[numvis * numhid + numvis:]).T
self.verbose = verbose
Contrastive.__init__(self, normalizeacrosscliques)
def __init__(self, normalizeacrosscliques=False):
Contrastive.__init__(self, normalizeacrosscliques)
def __init__(self,normalizeacrosscliques=False):
Contrastive.__init__(self,normalizeacrosscliques)