def run_atomic_light(len_str, len_alpha, nb_part, nb_run, extremum):
nb_different_motifs = []
nb_occurs = []
nb_occurs_avg = []
nb_occurs_sd = []
for _ in xrange(nb_run):
rstr = Rstr_max()
rs = random_string(len_str, len_alpha)
lp = cut_str(rs, nb_part)
for substring in lp:
rstr.add_str(substring)
r = rstr.go()
nb_different_motifs_run = len(r.keys())
nb_different_motifs.append(nb_different_motifs_run)
l_nb_occ = [nb for (_, nb) in r.keys()]
sum_nb_occ = sum(l_nb_occ)
nb_occurs_avg.append(float(sum_nb_occ) / nb_different_motifs_run)
nb_occurs_sd.append(tm.standart_deviation(l_nb_occ))
nb_occurs.append(sum_nb_occ)
if extremum > 0:
nb_different_motifs.sort()
nb_different_motifs = nb_different_motifs[extremum:-extremum]
nb_occurs.sort()
nb_occurs = nb_occurs[extremum:-extremum]
nb_occurs_avg.sort()
nb_occurs_avg = nb_occurs[extremum:-extremum]
# sd_motifs = tm.standart_deviation(nb_different_motifs)
# sd_occurs = tm.standart_deviation(nb_occurs)
moy = sum(nb_different_motifs) / len(nb_different_motifs)
moy_occur = sum(nb_occurs) / len(nb_occurs)
avg_occur = sum(nb_occurs_avg) / len(nb_occurs_avg)
sd_occur = sum(nb_occurs_sd) / len(nb_occurs_sd)
# return moy, moy_occur, avg_occur, sd_motifs, sd_occurs, dic_occ_len, dic_mot_len
return moy, moy_occur, avg_occur, sd_occur
def run_atomic_light(len_str, len_alpha, nb_part, nb_run, extremum) :
nb_different_motifs = []
nb_occurs = []
nb_occurs_avg = []
nb_occurs_sd = []
for _ in xrange(nb_run) :
rstr = Rstr_max()
rs = random_string(len_str,len_alpha)
lp = cut_str(rs,nb_part)
for substring in lp :
rstr.add_str(substring)
r = rstr.go()
nb_different_motifs_run = len(r.keys())
nb_different_motifs.append(nb_different_motifs_run)
l_nb_occ = [nb for (_, nb) in r.keys()]
sum_nb_occ = sum(l_nb_occ)
nb_occurs_avg.append(float(sum_nb_occ) / nb_different_motifs_run)
nb_occurs_sd.append(tm.standart_deviation(l_nb_occ))
nb_occurs.append(sum_nb_occ)
if extremum > 0 :
nb_different_motifs.sort()
nb_different_motifs = nb_different_motifs[extremum:-extremum]
nb_occurs.sort()
nb_occurs = nb_occurs[extremum:-extremum]
nb_occurs_avg.sort()
nb_occurs_avg = nb_occurs[extremum:-extremum]
# sd_motifs = tm.standart_deviation(nb_different_motifs)
# sd_occurs = tm.standart_deviation(nb_occurs)
moy = sum(nb_different_motifs) / len(nb_different_motifs)
moy_occur = sum(nb_occurs) / len(nb_occurs)
avg_occur = sum(nb_occurs_avg) / len(nb_occurs_avg)
sd_occur = sum(nb_occurs_sd) / len(nb_occurs_sd)
# return moy, moy_occur, avg_occur, sd_motifs, sd_occurs, dic_occ_len, dic_mot_len
return moy, moy_occur, avg_occur, sd_occur
#!/usr/bin/env python # -*- coding: utf-8 -*- import tool_math as tma rep = [0,1,2,3,4,5,6] print 'rep ::', rep av = tma.avg(rep) print 'av ::', av sd = tma.standart_deviation(rep) print 'sd ::', sd ev = tma.expected_value(rep) print 'ev ::', ev regr = tma.regression_lineaire_simple(rep) print 'regr ::',regr
def run_atomic(len_str, len_alpha, nb_part, nb_run, extremum):
nb_different_motifs = []
nb_occurs = []
nb_occurs_avg = []
nb_occurs_sd = []
stat_alpha = []
dic_zipf = {}
dic_occ_len = {}
dic_mot_len = {}
for _ in xrange(nb_run):
rstr = Rstr_max()
rs = random_string(len_str, len_alpha)
lp = cut_str(rs, nb_part)
for substring in lp:
rstr.add_str(substring)
r = rstr.go()
list_zipf = []
cpt = 0
for (_, nb), (l, _) in r.iteritems():
if l not in dic_occ_len:
dic_occ_len[l] = 0.
if l not in dic_mot_len:
dic_mot_len[l] = 0.
dic_occ_len[l] += float(nb) / nb_run
dic_mot_len[l] += 1. / nb_run
list_zipf.append(nb)
cpt += 1
list_zipf.sort()
for i in xrange(1, cpt + 1):
val = list_zipf[-i]
if i not in dic_zipf:
dic_zipf[i] = 0.
dic_zipf[i] += float(val) / nb_run
nb_different_motifs_run = len(r.keys())
nb_different_motifs.append(nb_different_motifs_run)
l_nb_occ = [nb for (_, nb) in r.keys()]
sum_nb_occ = sum(l_nb_occ)
nb_occurs_avg.append(float(sum_nb_occ) / nb_different_motifs_run)
nb_occurs_sd.append(tm.standart_deviation(l_nb_occ))
nb_occurs.append(sum_nb_occ)
if extremum > 0:
nb_different_motifs.sort()
nb_different_motifs = nb_different_motifs[extremum:-extremum]
nb_occurs.sort()
nb_occurs = nb_occurs[extremum:-extremum]
nb_occurs_avg.sort()
nb_occurs_avg = nb_occurs[extremum:-extremum]
# sd_motifs = tm.standart_deviation(nb_different_motifs)
# sd_occurs = tm.standart_deviation(nb_occurs)
moy = sum(nb_different_motifs) / len(nb_different_motifs)
moy_occur = sum(nb_occurs) / len(nb_occurs)
avg_occur = sum(nb_occurs_avg) / len(nb_occurs_avg)
sd_occur = sum(nb_occurs_sd) / len(nb_occurs_sd)
# return moy, moy_occur, avg_occur, sd_motifs, sd_occurs, dic_occ_len, dic_mot_len
return moy, moy_occur, avg_occur, sd_occur, dic_occ_len, dic_mot_len, dic_zipf
def run_atomic(len_str, len_alpha, nb_part, nb_run, extremum) :
nb_different_motifs = []
nb_occurs = []
nb_occurs_avg = []
nb_occurs_sd = []
stat_alpha = []
dic_zipf = {}
dic_occ_len= {}
dic_mot_len= {}
for _ in xrange(nb_run) :
rstr = Rstr_max()
rs = random_string(len_str,len_alpha)
lp = cut_str(rs,nb_part)
for substring in lp :
rstr.add_str(substring)
r = rstr.go()
list_zipf = []
cpt = 0
for (_,nb),(l,_) in r.iteritems() :
if l not in dic_occ_len :
dic_occ_len[l] = 0.
if l not in dic_mot_len :
dic_mot_len[l] = 0.
dic_occ_len[l] += float(nb) / nb_run
dic_mot_len[l] += 1. / nb_run
list_zipf.append(nb)
cpt += 1
list_zipf.sort()
for i in xrange(1,cpt+1) :
val = list_zipf[-i]
if i not in dic_zipf :
dic_zipf[i] = 0.
dic_zipf[i] += float(val) / nb_run
nb_different_motifs_run = len(r.keys())
nb_different_motifs.append(nb_different_motifs_run)
l_nb_occ = [nb for (_, nb) in r.keys()]
sum_nb_occ = sum(l_nb_occ)
nb_occurs_avg.append(float(sum_nb_occ) / nb_different_motifs_run)
nb_occurs_sd.append(tm.standart_deviation(l_nb_occ))
nb_occurs.append(sum_nb_occ)
if extremum > 0 :
nb_different_motifs.sort()
nb_different_motifs = nb_different_motifs[extremum:-extremum]
nb_occurs.sort()
nb_occurs = nb_occurs[extremum:-extremum]
nb_occurs_avg.sort()
nb_occurs_avg = nb_occurs[extremum:-extremum]
# sd_motifs = tm.standart_deviation(nb_different_motifs)
# sd_occurs = tm.standart_deviation(nb_occurs)
moy = sum(nb_different_motifs) / len(nb_different_motifs)
moy_occur = sum(nb_occurs) / len(nb_occurs)
avg_occur = sum(nb_occurs_avg) / len(nb_occurs_avg)
sd_occur = sum(nb_occurs_sd) / len(nb_occurs_sd)
# return moy, moy_occur, avg_occur, sd_motifs, sd_occurs, dic_occ_len, dic_mot_len
return moy, moy_occur, avg_occur, sd_occur, dic_occ_len, dic_mot_len, dic_zipf
