def test():
import sys
sys.path.insert(0,'digit_reconstructor\\rbm')
sys.path.insert(0,'digit_reconstructor\\dbn')
print("Loading DBN\n")
import cPickle as cP
with open("data\\defValues-300-300.pklb","rb") as f:
a=cP.load(f)
import dbn
import rbm0_m as rbm0
r=rbm0.rbm0()
r.weights=a[0]
r.load_data()
d=dbn.dbn([r],30)
d.rbmList[1].weights=a[1]
print("Loaded DBN\n")
epochs = 20
print("Training DBN\n")
d.train(epochs)
dweights =[]
print("Saving DBN weights\n")
for x in xrange(len(d.rbmList)):
dweights.append(d.rbmList[x].weights)
file_write_name="new_weights-784-300-30.pklb"
with open(file_write_name,"wb") as f:
cP.dump(dweights,f)
print("Displaying DBN reconstructed digits")
import dbninter
dbninter.display(d)
def __init__(self, rbmList, no_final_h, mrat=.5, eta=.001, p=.01):
sizes = [rbmList[0].size[0]]
sizes.extend([rbmList[k].size[1] for k in xrange(len(rbmList))])
sizes.extend([no_final_h])
self.rbmList = [rbmList[k] for k in xrange(len(rbmList))]
self.rbmList.append(rbm0.rbm0((sizes[-2], sizes[-1])))
self.rbmList[-1].data = rbmList[0].data
self.sizes = sizes
#initialize weights
#initalize network
self.data = rbmList[0].data
self.num_layers = len(rbmList)
self.default_weight_initializer()
self.reconError = 0 #sum of squared error terms, averaged over a mini_batch
#np.empty((785,0))#this is the distance between the trainV and the reconV, summed over a mini_batch
self.p = p # the desired sparsity of the hidden activations
self.q = 0
self.mrat = mrat
self.eta = eta
#Interpreter python2
#displays the dbn reconstructed digits
import sys
sys.path.insert(0,'..\\rbm')
print("Loading DBN\n")
import cPickle as cP
with open("..\\..\\data\\defValues-300-300.pklb","rb") as f:
a=cP.load(f)
import dbn
import rbm0_m as rbm0
r=rbm0.rbm0()
r.weights=a[0]
r.load_data()
d=dbn.dbn([r],30)
d.rbmList[1].weights=a[1]
print("Displaying DBN\n")
import dbninter
dbninter.display(d)
def defdisplay():
r = rbm0.rbm0()
r.load_binary_images()
import cPickle as cP
r.weights = cP.load(open("rbmWeights_300h.pklb", "rb"))
display(r)