def get_rrxor_tree():
rr = RRXOR()
rstr = rr.simulate(1000000*3)
bpt = BinaryParseTree(3, 2, 1)
bpt.parse(rstr)
return bpt
def test_golden_mean():
gm = GoldenMean()
gstr = gm.simulate(500000)
for L in range(2, 5):
expected_words = get_all_binary_words(L)
ewdist = empirical_word_distributions(gstr, L, expected_words)
awdist = analytical_word_distributions(gm, L)
dprobs = dict()
for ws,wp in ewdist.iteritems():
dprobs[ws] = list()
dprobs[ws].append(wp)
for ws,wp in awdist.iteritems():
dprobs[ws].append(wp)
print 'Distribution of length %d (empirical, analytical)' % L
for ws,(ewp,awp) in dprobs.iteritems():
print '%s: %0.4f, %0.4f' % (ws, ewp, awp)
Lvals = range(1, 8)
plot_process_info(gm, Lvals)
bpt = BinaryParseTree(5, 2, 2)
bpt.parse(gstr)
bpt.show()
def get_evenprocess_tree(dthresh=1e-1, D=5, M=2, L=3):
ep = EvenProcess()
estr = ep.simulate(1000000)
bpt = BinaryParseTree(D, M, L, dthresh=dthresh)
bpt.parse(estr)
return bpt
def get_rrxor_tree():
rr = RRXOR()
rstr = rr.simulate(1000000 * 3)
bpt = BinaryParseTree(3, 2, 1)
bpt.parse(rstr)
return bpt
def get_gm_tree():
gm = GoldenMean()
gstr = gm.simulate(500000)
bpt = BinaryParseTree(5, 2, 2)
bpt.parse(gstr)
return bpt
def get_evenprocess_tree(dthresh=1e-1, D=5, M=2, L=3):
ep = EvenProcess()
estr = ep.simulate(1000000)
bpt = BinaryParseTree(D, M, L, dthresh=dthresh)
bpt.parse(estr)
return bpt
def get_gm_tree():
gm = GoldenMean()
gstr = gm.simulate(500000)
bpt = BinaryParseTree(5, 2, 2)
bpt.parse(gstr)
return bpt
def test_biasedcoin(p=0.5):
bc = BiasedCoin(p=p)
bstr = bc.simulate(100000)
bpt = BinaryParseTree(4, 2, 4)
bpt.parse(bstr)
bpt.show()
def get_gee_tree(dthresh=1e-3):
wdist = {
'00100':1.0/8.0,
'00110':1.0/16.0,
'00111':1.0/16.0,
'01001':1.0/8.0,
'01100':1.0/16.0,
'01110':1.0/32.0,
'01111':1.0/32.0,
'10010':1.0/8.0,
'10011':1.0/8.0,
'11001':1.0/8.0,
'11100':1.0/16.0,
'11110':1.0/32.0,
'11111':1.0/32.0
}
bpt = BinaryParseTree(5, 2, 2, dthresh=dthresh)
bpt.parse_from_word_distribution(wdist)
return bpt
def get_gee_tree(dthresh=1e-3):
wdist = {
'00100': 1.0 / 8.0,
'00110': 1.0 / 16.0,
'00111': 1.0 / 16.0,
'01001': 1.0 / 8.0,
'01100': 1.0 / 16.0,
'01110': 1.0 / 32.0,
'01111': 1.0 / 32.0,
'10010': 1.0 / 8.0,
'10011': 1.0 / 8.0,
'11001': 1.0 / 8.0,
'11100': 1.0 / 16.0,
'11110': 1.0 / 32.0,
'11111': 1.0 / 32.0
}
bpt = BinaryParseTree(5, 2, 2, dthresh=dthresh)
bpt.parse_from_word_distribution(wdist)
return bpt
def test_golden_mean():
gm = GoldenMean()
gstr = gm.simulate(500000)
for L in range(2, 5):
expected_words = get_all_binary_words(L)
ewdist = empirical_word_distributions(gstr, L, expected_words)
awdist = analytical_word_distributions(gm, L)
dprobs = dict()
for ws, wp in ewdist.iteritems():
dprobs[ws] = list()
dprobs[ws].append(wp)
for ws, wp in awdist.iteritems():
dprobs[ws].append(wp)
print 'Distribution of length %d (empirical, analytical)' % L
for ws, (ewp, awp) in dprobs.iteritems():
print '%s: %0.4f, %0.4f' % (ws, ewp, awp)
Lvals = range(1, 8)
plot_process_info(gm, Lvals)
bpt = BinaryParseTree(5, 2, 2)
bpt.parse(gstr)
bpt.show()
def test_biasedcoin(p=0.5):
bc = BiasedCoin(p=p)
bstr = bc.simulate(100000)
bpt = BinaryParseTree(4, 2, 4)
bpt.parse(bstr)
bpt.show()
def get_oh_tree(dthresh=1e-3):
wdist = {
'00000':1.0/40.0,
'00001':1.0/40.0,
'00010':1.0/40.0,
'00011':1.0/40.0,
'00100':1.0/40.0,
'00101':1.0/40.0,
'00111':1.0/20.0,
'01000':1.0/40.0,
'01001':1.0/40.0,
'01010':1.0/40.0,
'01011':1.0/40.0,
'01110':1.0/20.0,
'01111':1.0/20.0,
'10000':1.0/40.0,
'10001':1.0/40.0,
'10010':1.0/40.0,
'10011':1.0/40.0,
'10100':1.0/40.0,
'10101':1.0/40.0,
'10111':1.0/20.0,
'11000':1.0/40.0,
'11001':1.0/40.0,
'11010':1.0/40.0,
'11011':1.0/40.0,
'11100':1.0/20.0,
'11101':1.0/20.0,
'11110':1.0/20.0,
'11111':3.0/20.0
}
bpt = BinaryParseTree(5, 2, 2, dthresh=dthresh)
bpt.parse_from_word_distribution(wdist)
return bpt
def get_oh_tree(dthresh=1e-3):
wdist = {
'00000': 1.0 / 40.0,
'00001': 1.0 / 40.0,
'00010': 1.0 / 40.0,
'00011': 1.0 / 40.0,
'00100': 1.0 / 40.0,
'00101': 1.0 / 40.0,
'00111': 1.0 / 20.0,
'01000': 1.0 / 40.0,
'01001': 1.0 / 40.0,
'01010': 1.0 / 40.0,
'01011': 1.0 / 40.0,
'01110': 1.0 / 20.0,
'01111': 1.0 / 20.0,
'10000': 1.0 / 40.0,
'10001': 1.0 / 40.0,
'10010': 1.0 / 40.0,
'10011': 1.0 / 40.0,
'10100': 1.0 / 40.0,
'10101': 1.0 / 40.0,
'10111': 1.0 / 20.0,
'11000': 1.0 / 40.0,
'11001': 1.0 / 40.0,
'11010': 1.0 / 40.0,
'11011': 1.0 / 40.0,
'11100': 1.0 / 20.0,
'11101': 1.0 / 20.0,
'11110': 1.0 / 20.0,
'11111': 3.0 / 20.0
}
bpt = BinaryParseTree(5, 2, 2, dthresh=dthresh)
bpt.parse_from_word_distribution(wdist)
return bpt
