def func():
all_cas_d = load_obj('C:\weibodataset\\2011_08c_ind.pkl')
outnei_d = load_obj(ON_PATH)
exp = outnei_d.keys()[0]
print exp,type(exp)
comm_list = comm_l()
len_comm_l = len(comm_list) - 1
result_d = {}
for rtmid in all_cas_d:
if random.random() <= 0.9:
continue
rtmid_d = all_cas_d[rtmid]
onexp_d = {} #onexp_d[outnei_id] ---> [adopters who are the outnei's active incoming neighbors]
for adp_ind in rtmid_d:
#print adp_ind,type(adp_ind)
#raise
if int(adp_ind) <= len_comm_l:
comm_no_adp = comm_list[int(adp_ind)]
else:
comm_no_adp = -int(adp_ind) #singleton comm added to match V1
adp_ind_str = str(adp_ind).replace('L','')
if adp_ind_str in outnei_d:
outneis = outnei_d[adp_ind_str]
for outnei in outneis:
if outnei in onexp_d:
onexp_d[outnei].append(comm_no_adp)
else:
onexp_d[outnei] = [comm_no_adp]
#print onexp_d
for k,v in onexp_d.iteritems():
an = len(v)
ac = len(set(v))
key = str(an)+','+str(ac)
if key not in result_d:
result_d[key] = 1
else:
result_d[key] += 1
pass
save_obj(result_d,RESULT_PATH)
def func(pertss,cas_cent_path=cas_cent_path):
fns = os.listdir(cas_cent_path)
print fns
for_hm = []
for i in xrange(0,len(fns)):
fn = fns[i]
full_path = os.path.join(cas_cent_path,fn)
cas_cent_list = load_obj(full_path)
perts = pertss[i]
for_hm_cent = [] # = [[# of nodes over thres1, #of nodes over thres2, # of nodes over thres3, # of nodes over thres4...]]
for one_cas in cas_cent_list:
temp_l = []
for i in xrange(0,5):
cent_thres = perts[i]
#result = []
nodes_over_thres = [x for x in one_cas if x >= cent_thres]
number_nodes_over_thres = len(nodes_over_thres)
#for_hm_cent.append()
temp_l.append(number_nodes_over_thres)
for_hm_cent.append(temp_l)
for_hm.append(heatmap(for_hm_cent))
return for_hm
def func(CKs):
print 'Choose the time as %d, with lambda equal to %d' %(THRES_TIME,LAMBDA)
#CK = 'com_k_shell'
#print 'current centrality key is '+CK
core_d_outnei = load_obj(CORE_PATH)
#cm_d = comm_dict()
#comm_dist = {}
#comm_dist['size'] = THRES_SIZE
#comm_dist['lbd'] = LAMBDA
ind = 0
#cent_list, perts = load_cents(CK)
#perts = np.percentile(cent_list,pert_thres)
of = open('E:\\SNAM_2015\\time_cent_percentiles.csv','wb')
writer = csv.writer(of)
titles = ['ind']
for k in xrange(0,len(CKs)):
ck = CKs[k]
for j in xrange(0,len(pert_thres)):
pert = str(pert_thres[j])
title = ck+'_'+pert
titles.append(title)
titles.append(ck+'_avg')
writer.writerow(titles)
all_cent_lists = []
pertss = []
for CK in CKs:
cent_list,perts = load_cents(CK)
all_cent_lists.append(cent_list)
pertss.append(perts)
for rtmid in core_d_outnei:
rtmid_d = core_d_outnei[rtmid]
adps = []
final_size = len(rtmid_d.keys())
ind += 1
#if final_size< THRES_SIZE:
# continue
#tfr_list = [] # find the first THRES_SIZE adopters
for adp_ in rtmid_d:
t_diff = abs(int(rtmid_d[adp_][0]))+0.01*random.random()
if t_diff <= THRES_TIME:
adps.append(adp_)
'''
tfr_dict = {}
for adp_ in rtmid_d:
t_diff = abs(int(rtmid_d[adp_][0]))+0.01*random.random()
#print 'time from root for this adp %d is %d, t_diif - T_FR = %d' %(int(adp_),t_diff,t_diff-T_FR)
tfr_dict[int(adp_)] = t_diff
tfr_list = tfr_dict.values()
tfr_list.sort()
#t_time = tfr_list[THRES_SIZE-1] - tfr_list[0]
#avg_t_fr = float(sum(tfr_list[0:(THRES_SIZE-1)]))/THRES_SIZE #average time for one adoption behavior happen, not average exposure time for adopters
for adp_ in tfr_dict:
t_diff = tfr_dict[adp_]
if tfr_list.index(t_diff) < THRES_SIZE:
adps.append(adp_)
'''
#For the size version, we also need a time stamp to judge if the outnei is a NA or FNT
#So I pick the time when the 50th adopter reposted the msg as time stamp
#T_FR = tfr_list[THRES_SIZE-1]
#cur_size = len(adps)
set_adps = set(adps)
'''
cur_size = len(set_adps)
if cur_size != THRES_SIZE:
print cur_size
raise 'the size is not correct'
'''
row = [ind]
for x in xrange(0,len(all_cent_lists)):
cent_list = all_cent_lists[x]
perts = pertss[x]
cent_vs = []
for adp in set_adps:
cent_vs.append(cent_list[adp])
avg_cent_v = np.average(cent_vs)
num_nodes_over_perts = []
for i in xrange(0,len(perts)):
num_nodes_over_perts.append(0)
for v in cent_vs:
for j in xrange(0,len(perts)):
if v > perts[j]:
num_nodes_over_perts[j] += 1
num_nodes_over_perts.append(avg_cent_v)
row.extend(num_nodes_over_perts)
#print num_nodes_over_perts,ind
writer.writerow(row)
for_hm_cent.append(temp_l)
for_hm.append(heatmap(for_hm_cent))
return for_hm
if __name__ == '__main__':
pertss = []
fns = os.listdir(cas_cent_path)
#print fns
for_hm = []
for fn in fns:
full_path = os.path.join(cas_cent_path,fn)
cas_cent_list = load_obj(full_path)
perts = list(percentiles(cas_cent_list))
pertss.append(perts)
hm_all = func(pertss,cas_cent_path=cas_cent_path)
hm_viral = func(pertss,cas_cent_path=cas_cent_path+'_v')
for i in xrange(0,len(os.listdir(cas_cent_path))):
hm_cent = hm_all[i].reshape(-1)
hm_cent_v = hm_viral[i].reshape(-1)
result = []
for i in xrange(0,len(hm_cent)):
deno = hm_cent[i]
enum = hm_cent_v[i]
if deno == 0:
result.append(0)
__author__ = "rguo12"
path = "C:\\weibodataset\\heat_map_dv1_notact.pkl"
from asonam_caspre.tool_box import load_obj, save_obj
import matplotlib.pyplot as plt
import numpy as np
if __name__ == "__main__":
d = load_obj(path)
result = []
for i in xrange(1, 6):
row = []
for j in xrange(1, 6):
key = str(i) + "," + str(j)
if key in d:
row.append(d[key])
else:
row.append(0)
result.append(row)
print result
result = np.asarray(result)
"""
fig, ax = plt.subplots()
plt.subplots_adjust(left=0.2, right=0.95, top=0.95, bottom=0.2)
heatmap = ax.pcolor(result, cmap=plt.cm.Blues)
cbar = plt.colorbar(heatmap)
ax.tick_params(axis='both', which='major', labelsize=24)