rms2 = []
nbEpoch = 60
l_ex = list(range(80))
shuffle(l_ex)
ptrain_pattern_l = l_ex[0:40]
#pre-training
for epoch in range(nbEpoch):
l_exx = list(range(80))
shuffle(l_exx)
rms_ss = 0.
rms_ss2 = 0.
for ex in l_exx:
first_order.calc_output(ptrain_pattern[ex][0])
#compara
compara = []
for i in range(48):
compara.append(ptrain_pattern[ex][0][i] -
first_order.stateOutputNeurons[i])
res2 = [high_order[i].calc_output(compara) for i in range(2)]
if (index_max(res2) == 0):
err += 1
if (pattern_to_list(
first_order.stateOutputNeurons) == pattern_to_list(
ptrain_pattern[ex][1])):
res = [high_order[i].calc_output(compara) for i in range(2)]
rms2 = []
nbEpoch = 60
l_ex = list(range(80))
shuffle(l_ex)
ptrain_pattern_l = l_ex[0:40]
# pre-training
for epoch in range(nbEpoch):
l_exx = list(range(80))
shuffle(l_exx)
rms_ss = 0.0
rms_ss2 = 0.0
for ex in l_exx:
first_order.calc_output(ptrain_pattern[ex][0])
# compara
compara = []
for i in range(48):
compara.append(ptrain_pattern[ex][0][i] - first_order.stateOutputNeurons[i])
res2 = [high_order[i].calc_output(compara) for i in range(2)]
if index_max(res2) == 0:
err += 1
if pattern_to_list(first_order.stateOutputNeurons) == pattern_to_list(ptrain_pattern[ex][1]):
res = [high_order[i].calc_output(compara) for i in range(2)]
if index_max(res) == 0:
rms_ss += 1
