def demo():
parser = argparse.ArgumentParser()
parser.add_argument('-f',
action='store',
required=True,
dest='trainfile',
help='file for train')
parser.add_argument('-w',
action='store',
required=True,
dest='weightfile',
help='weight file for train')
inputs = parser.parse_args()
writeTrainWeights(inputs.weightfile)
pointNum, pat = readfile(learnParameters.expandfeature(inputs.trainfile))
print pat
print('file read over ,the train has began please wait for a momment')
n = NN(pointNum, pointNum, 1)
#for i in xrange(100):
f_wfile = open(inputs.weightfile, 'r')
for line in f_wfile:
line = line.strip('\n')
line = line.split(':')
iterations = int(line[0])
N_value = float(line[1])
M_value = float(line[2])
print('iterations=%d,N=%f,M=%f' % (iterations, N_value, M_value))
n.train(pat, iterations, N_value, M_value)
n.weights()
print('Next paratermeters........')
f_wfile.close()
def demo():
parser=argparse.ArgumentParser()
parser.add_argument('-f', action='store', required=True, dest='trainfile',
help='file for train')
parser.add_argument('-w', action='store', required=True, dest='weightfile',
help='weight file for train')
inputs=parser.parse_args()
writeTrainWeights(inputs.weightfile)
pointNum,pat=readfile(learnParameters.expandfeature(inputs.trainfile))
print pat
print('file read over ,the train has began please wait for a momment')
n=NN(pointNum,pointNum,1)
#for i in xrange(100):
f_wfile=open(inputs.weightfile,'r')
for line in f_wfile:
line=line.strip('\n')
line=line.split(':')
iterations=int(line[0])
N_value=float(line[1])
M_value=float(line[2])
print('iterations=%d,N=%f,M=%f'%(iterations,N_value,M_value))
n.train(pat,iterations,N_value,M_value)
n.weights()
print('Next paratermeters........')
f_wfile.close()
def psolearn(trainfile, N_value, M_value):
pointNum, pat = readfile(learnParameters.expandfeature(trainfile))
print('file read over ,the train has began please wait for a momment')
n = NN(pointNum, pointNum, 1)
#for i in xrange(100):
iterations = 1000
value = n.train(pat, iterations, N_value, M_value)
print('N_value:%f,M_value:%f' % (N_value, M_value))
#n.weights()
return value
def psolearn(trainfile,N_value,M_value):
pointNum,pat=readfile(learnParameters.expandfeature(trainfile))
print('file read over ,the train has began please wait for a momment')
n=NN(pointNum,pointNum,1)
#for i in xrange(100):
iterations=1000
value=n.train(pat,iterations,N_value,M_value)
print ('N_value:%f,M_value:%f'%(N_value,M_value))
#n.weights()
return value