def draw_graph():
#user participation
user_part_num = [len(rumor_num) for rumor_num in user_participation.values()]
cdf = CDFPlot()
cdf.set_label('Number of rumors', 'CDF')
cdf.set_data(user_part_num, 'CDF')
cdf.save_image('Image/%s/user_participation_cdf.png'%foldername)
ccdf = CCDFPlot()
ccdf.set_label('Number of rumors', 'CCDF')
ccdf.set_data(user_part_num)
ccdf.save_image('Image/%s/user_participation_ccdf.png'%foldername)
top_participated_users(user_part_num)
def sharecount_cdf(date_condition, file_name):
sql = """
SELECT post_id, share_count
FROM snopes_set
WHERE published_date %s
""" % date_condition
cursor.execute(sql)
rs = cursor.fetchall()
sharecount_list = []
for item in rs:
post_id, share_count, = item
sharecount_list.append(int(share_count))
#Display CDF and save at the path
Cdf = CDFPlot()
Cdf.set_label('share_count', 'CDF')
Cdf.set_log(True)
Cdf.set_data(sharecount_list, "")
Cdf.save_image('./image/sharecount_%s.png' % file_name)
return sharecount_list
def depth_cdf():
depth = list(itertools.chain(*list(itertools.chain(*[item.values() for item in get_depth('True').values()]))))
#depth = list(itertools.chain(*depth))
depth2 = list(itertools.chain(*list(itertools.chain(*[item.values() for item in get_depth('False').values()]))))
depth3 = list(itertools.chain(*list(itertools.chain(*[item.values() for item in get_depth('Mixture,Mostly False,Mostly True').values()]))))
cdf = CDFPlot()
cdf.set_label('Depth', 'CDF')
cdf.set_log(True)
cdf.set_ylog()
cdf.set_data(depth, 'True')
cdf.set_data(depth2, 'False')
cdf.set_data(depth3, 'Mixed')
cdf.set_legends(['True', 'False', 'Mixed'], '')
cdf.save_image('Image/depth_cdf.png')
def max_depth_per_rumor():
depth = []; depth2 = []; depth3 = []
for item in get_depth('True').values():
for t in item.values():
depth.append(max(t))
for item in get_depth('False').values():
for t in item.values():
depth2.append(max(t))
for item in get_depth('Mixture,Mostly False,Mostly True').values():
for t in item.values():
depth3.append(max(t))
#cdf_plot.set_label_num(3)
cdf = CDFPlot()
cdf.set_label('Cascade Max Depth', 'CDF')
cdf.set_log(True)
cdf.set_ylog()
cdf.set_data(depth, 'True')
cdf.set_data(depth2, 'False')
cdf.set_data(depth3, 'Mixed')
cdf.set_legends(['True', 'False', 'Mixed'], '')
cdf.save_image('Image/depth_per_cascade_cdf.png')
for i in range(1, 20):
print(i, depth.count(i), depth2.count(i), depth3.count(i))
def cascade_num():
cascades = [len(item) for item in get_cascades('True').values()]
#print(cascades)
cascades2 = [len(item) for item in get_cascades('False').values()]
cascades3 = [
len(item)
for item in get_cascades('Mixture,Mostly False,Mostly True').values()
]
print(len(cascades), len(cascades2), len(cascades3))
cdf_plot.set_label_num(2)
cdf = CDFPlot()
#cdf.set_label('Number of Cascades', 'CDF')
cdf.set_log(True)
cdf.set_ylog()
cdf.set_data(cascades, 'True')
cdf.set_data(cascades2, 'False')
cdf.set_data(cascades3, 'Mixed')
cdf.set_legends(['True', 'False', 'Mixed'], '')
cdf.save_image('Image/cascades_number_cdf.png')
def cascade_cdf():
cascades = get_cascades('True')
cascades = list(
itertools.chain(
*[item.values() for item in get_cascades('True').values()]))
#cascades = list(itertools.chain(*cascades))
cascades2 = list(
itertools.chain(
*[item.values() for item in get_cascades('False').values()]))
cascades3 = list(
itertools.chain(*[
item.values() for item in get_cascades(
'Mixture,Mostly False,Mostly True').values()
]))
cdf = CDFPlot()
cdf.set_label('Cascade Size', 'CDF')
cdf.set_log(True)
cdf.set_ylog()
cdf.set_data(cascades, 'True')
cdf.set_data(cascades2, 'False')
cdf.set_data(cascades3, 'Mixed')
cdf.set_legends(['True', 'False', 'Mixed'], '')
cdf.save_image('Image/cascades_cdf.png')
def rumor_statistics():
rumor_unique_cascade = {}
rumor_cascade_size = {}
for ccc, postid in enumerate(rumors):
#if postid != '126119':
# continue
rumor_unique_cascade[postid] = {}
with open(dirname + '/' + postid, 'r') as f:
tweets = json.load(f)
for tweet in tweets.values():
utid = tweet['origin_tweet']
rumor_unique_cascade[postid][utid] = tweet['cascade']
#if ccc == 10:
# break
#for item in rumor_unique_cascade.values():
# print(len(item))
rumor_nums = [len(item) for item in rumor_unique_cascade.values()]
rumor_size = [np.mean(item.values()) for item in rumor_unique_cascade.values()]
rumor_size_all = list(itertools.chain(*[item.values() for item in rumor_unique_cascade.values()]))
#depth = list(itertools.chain(*list(itertools.chain(*[item.values() for item in get_depth('True').values()]))))
#print(rumor_nums)
#print(rumor_size_all)
cdf = CDFPlot()
cdf.set_label('Number of Cascade', 'CDF')
cdf.set_title('Distribution of Mean Number of Cascades per Rumor')
cdf.set_log(True)
cdf.set_data(rumor_nums, '')
cdf.save_image("Image/%s/cascade_num_per_rumor_cdf.png"%(foldername))
cdf = CCDFPlot()
cdf.set_log(True)
cdf.set_data(rumor_nums)
cdf.save_image("Image/%s/cascade_num_per_urmor_ccdf.png"%(foldername))
cdf = CDFPlot()
cdf.set_label('Cascades Size', 'CDF')
cdf.set_title('Distribution of Mean Size of Cascades of per Rumor')
cdf.set_log(True)
cdf.set_data(rumor_size, '')
cdf.save_image("Image/%s/cascade_size_per_rumor_cdf.png"%(foldername))
cdf = CCDFPlot()
cdf.set_log(True)
cdf.set_data(rumor_size)
cdf.save_image("Image/%s/cascade_size_per_urmor_ccdf.png"%(foldername))
cdf = CDFPlot()
cdf.set_label('Cascade Size', 'CDF')
cdf.set_title('Distribution of Cascade Size')
cdf.set_log(True)
cdf.set_data(rumor_size_all, '')
cdf.save_image("Image/%s/cascade_size_all_rumor_cdf.png"%(foldername))
cdf = CCDFPlot()
cdf.set_label('Size of Cascade', 'CCDF')
cdf.set_log(True)
cdf.set_data(rumor_size_all)
cdf.save_image("Image/%s/cascade_size_all_rurmor_ccdf.png"%(foldername))
def draw_cdf_plot(datas, datatype, legend, legend_type, filename, log_scale=True):
cdf = CDFPlot()
cdf.set_label(datatype, 'CDF')
cdf.set_log(log_scale)
for i in range(len(datas)):
cdf.set_data(datas[i], legend[i])
cdf.set_xticks([-1, 0, 1],index = [-1, 0, 1])
if len(legend) > 1:
cdf.set_legends(legend, legend_type)
cdf.save_image('Image/%s/%s.png'%(foldername, filename))
def echo_chamber_diversity(filename):
Bot = bot.load_bot()
dirname = 'Retweet/'
files = os.listdir(dirname)
if filename == None:
echo_chamber_users = {}
for postid in files:
echo_chamber_users[postid] = {}
else:
echo_chamber_users = e_util.get_echo_chamber_users(filename)
echo_tweet_diversity = []; echo_source_diversity = [];
necho_tweet_diversity = []; necho_source_diversity = [];
for postid in files:
with open(dirname + postid) as f:
tweets = json.load(f)
non_echo_users = {}
for tweet in tweets.values():
user = tweet['user']
#non echo chamber collect
if not user in echo_chamber_users[postid]:
non_echo_users[user] = 1
print(len(echo_chamber_users[postid]), len(non_echo_users))
timeline_dir = '../Timeline/'
#collect echo chamber users' source diversity
err = 0; nerr = 0
for user in echo_chamber_users[postid]:
try:
with open(timeline_dir + user, 'r') as f:
user_tweets = json.load(f)
except IOError as e:
#print(e)
err +=1
continue
tweet_diversity, source_diversity = get_diversity(user_tweets)
if tweet_diversity != None:
echo_tweet_diversity.append(tweet_diversity)
if source_diversity != None:
echo_source_diversity.append(source_diversity)
for user in non_echo_users:
try:
with open(timeline_dir + user, 'r') as f:
user_tweets = json.load(f)
except IOError as e:
#print(e)
nerr += 1
continue
tweet_diversity, source_diversity = get_diversity(user_tweets)
if tweet_diversity != None:
necho_tweet_diversity.append(tweet_diversity)
if source_diversity != None:
necho_source_diversity.append(source_diversity)
#print(err, nerr)
#break
#CDF
cdf = CDFPlot()
cdf.set_label('Retweet Origin Diversity', 'CDF')
#cdf.set_log(True)
cdf.set_data(echo_tweet_diversity, 'Echo Chamber')
cdf.set_data(necho_tweet_diversity, 'Non Echo Chamber')
cdf.set_legends(['Echo CHamber', 'Non Echo CHamber'], 'User Type')
cdf.save_image('Image/20181002/source_diversity_retweet_cdf.png')
cdf = CDFPlot()
cdf.set_label('Source News Diversity', 'CDF')
#cdf.set_log(True)
cdf.set_data(echo_source_diversity, 'Echo Chamber')
cdf.set_data(necho_source_diversity, 'Non Echo Chamber')
cdf.set_legends(['Echo CHamber', 'Non Echo CHamber'], 'User Type')
cdf.save_image('Image/20181002/source_diversity_news_cdf.png')
#BoxPlot
box = BoxPlot(1)
box.set_data([echo_tweet_diversity, necho_tweet_diversity],'')
box.set_xticks(['Echo Chamber', 'Non Echo Chamber', 'All'])
box.set_label('', 'Retweet Origin Diversity')
box.save_image('Image/20181002/source_diversity_retweet.png')
box = BoxPlot(1)
box.set_data([echo_source_diversity, necho_source_diversity],'')
box.set_xticks(['Echo Chamber', 'Non Echo Chamber', 'All'])
box.set_label('', 'Source News Diversity')
box.save_image('Image/20181002/source_diversity_news.png')
def draw_graph():
depth_time1, depth_user1, unique_user_time1, cascade_depth1 = time_series('True')
x_ticks1 = depth_time1.keys()
y_ticks1 = [np.mean(depth_time1[depth].values()) for depth in x_ticks1]
depth_time2, depth_user2, unique_user_time2, cascade_depth2 = time_series('False')
x_ticks2 = depth_time2.keys()
y_ticks2 = [np.mean(depth_time2[depth].values()) for depth in x_ticks1]
#draw mean minutes - depth line plot
line = LinePlot()
line.set_ylog()
line.set_label('Depth', 'Mean Minutes')
line.set_plot_data([y_ticks1, y_ticks2], x_ticks1)
line.set_legends(['True', 'False'])
line.save_image('Image/time_depth_line.png')
x_ticks1 = unique_user_time1.keys()
x_ticks2 = unique_user_time2.keys()
x_ticks1 = sorted(x_ticks1)
y_ticks1 = [np.mean(unique_user_time1[num].values()) for num in x_ticks1]
y_ticks2 = [np.mean(unique_user_time2[num].values()) for num in x_ticks2]
#draw mean minutes - unique users line plot
line = LinePlot()
line.set_ylog()
line.set_label('Unique Users', 'Mean Minutes')
line.set_plot_data([y_ticks1, y_ticks2], x_ticks1)
line.set_xticks(x_ticks1)
line.set_legends(['True', 'False'])
line.save_image('Image/time_users_line.png')
all_depth_true = [[key] * len(depth_time1[key]) for key in depth_time1.keys()] #True
all_depth_false = [[key] * len(depth_time2[key]) for key in depth_time2.keys()] #True
all_depth_sum_true = []
all_depth_sum_false = []
for item in all_depth_true:
all_depth_sum_true.extend(item)
for item in all_depth_false:
all_depth_sum_false.extend(item)
#Depth CDF, CCDF
#cdf = CDFPlot()
#cdf.set_data(all_depth_sum_true, 'True')
#cdf.set_data(all_depth_sum_false, 'False')
#cdf.set_legends(['True', 'False'], '')
#cdf.save_image('Image/depth_cdf.png')
true_cascade = []
false_cascade = []
for postid in cascade_depth1.keys():
for depth in cascade_depth1[postid].values(): #origin tweet : depth
true_cascade.extend(depth)
for postid in cascade_depth2.keys():
for depth in cascade_depth2[postid].values(): #origin tweet : depth
false_cascade.extend(depth)
print('true')
for i in range(1, 15):
print(i, true_cascade.count(i))
print('false')
for i in range(1, 15):
print(i, false_cascade.count(i))
cdf = CDFPlot()
cdf.set_legends(['True', 'False'], '')
cdf.set_xlim(0, 11)
#cdf.set_log(True)
#cdf.set_ylog()
cdf.set_label('Depth', 'CDF')
cdf.set_data(true_cascade, 'True')
cdf.set_data(false_cascade, 'False')
cdf.save_image('Image/depth_cdf.png')
cursor.execute(sql)
rs = cursor.fetchall()
return [item[0] for item in rs]
if __name__ == "__main__":
#numberof articles
total_count = 0
#sql connect
conn, cursor, = sql_connect()
total_list = sharecount_cdf("<= date(now())", "total")
sub_total_list = sharecount_cdf("< '2018-03-01'", "2018_02")
Cdf = CDFPlot()
Cdf.set_label('share_count', 'CDF')
Cdf.set_log(True)
Cdf.set_data(total_list, "total")
Cdf.set_data(sub_total_list, "< 2018.02")
Cdf.save_image('./image/sharecount_%s.png' % "comparison")
sharecount_cdf("between '2016-01-01' and '2018-03-01'", "2016_2018")
#share count per year
year = [
"'2010-01-01' and '2010-12-31'", "'2011-01-01' and '2011-12-31'",
"'2012-01-01' and '2012-12-31'", "'2013-01-01' and '2013-12-31'",
"'2014-01-01' and '2014-12-31'", "'2015-01-01' and '2015-12-31'",
"'2016-01-01' and '2016-12-31'", "'2017-01-01' and '2017-12-31'"
]
BarPlt = BarPlot(1)
BarPlt.set_data(np.arange(max_num), count_list, "")
BarPlt.set_width(0.8)
BarPlt.set_xticks(np.arange(max_num))
BarPlt.save_image("./image/source_num_bar.png")
count_list = [source_count_list_from_2017.count(i) for i in range(max_num)]
BarPlt = BarPlot(1)
BarPlt.set_data(np.arange(max_num), count_list, "")
BarPlt.set_width(0.8)
BarPlt.set_xticks(np.arange(max_num))
BarPlt.save_image("./image/source_num_bar_2017.png")
#number of source distribution - cdf
Cdf = CDFPlot()
Cdf.set_label('number of sources', 'CDF')
Cdf.set_log(True)
Cdf.set_data(source_count_list, "")
Cdf.set_data(source_count_list_from_2017, "")
Cdf.set_legends(["All", "year >=2017"])
Cdf.save_image('./image/source_num_distribution_cdf.png')
#number of articles distirbution by source - cdf
veracity_list = veracity_types()
sources_json, veracities, postids = sources_count()
veracity_dict = {}
for v in veracity_list:
veracity_dict[v.lower()] = []
def echo_chamber_group_homogeneity_size():
filename = 'Data/echo_chamber2.json'
#echo_chamber_users = e_util.get_echo_chamber_users(filename)
f = open(filename, 'r')
echo_chamber = json.load(f)
f.close()
d = []
all_similarity = []
for ccc, key in enumerate(echo_chamber):
users = echo_chamber[key]
user_size = len(users)
if ccc % 100 == 0:
print(ccc)
if user_size < 2:
continue
similarities = []
for i in range(len(users)):
users1 = users[i]
p1 = get_polarity(users1)
for j in range(i +1, len(users)):
users2 = users[j]
p2 = get_polarity(users2)
if p1 == -999 or p2 == -999:
continue
similarity = p1 * p2
similarities.append(similarity)
all_similarity.append(round(similarity,2))
d.append({'size' : user_size, 'polarity' : round(np.median(similarity),2)})
size_list = [item['size'] for item in d]
polarity_list = [item['polarity'] for item in d]
#print(size_list)
scatter = ScatterPlot()
#scatter.set_log(True)
scatter.set_xlim(10000)
scatter.set_ylim(-1, 1.2)
scatter.set_data(size_list, polarity_list)
scatter.save_image('Image/%s/echo_chamber_polarity_size.png'%foldername)
cdf = CDFPlot()
cdf.set_data(all_similarity,'')
cdf.set_data(polarity_list,'')
cdf.set_label('Polarity', 'CDF')
cdf.set_legends(['All', 'Median'], '')
cdf.save_image('Image/%s/echo_chamber_all_polarity_similarity_cdf.png'%foldername)
def depth_politics_cdf():
depth, echo_depth, necho_depth = get_depth(politic=True, veracity='False', echo_chamber=True)
depth2, echo_depth2, necho_depth2 = get_depth(politic=False, veracity='False', echo_chamber=True)
"""
depth = list(itertools.chain(*list(itertools.chain(*[item.values() for item in depth.values()]))))
echo_depth = list(itertools.chain(*list(itertools.chain(*[item.values() for item in echo_depth.values()]))))
necho_depth = list(itertools.chain(*list(itertools.chain(*[item.values() for item in necho_depth.values()]))))
depth2 = list(itertools.chain(*list(itertools.chain(*[item.values() for item in depth2.values()]))))
echo_depth2 = list(itertools.chain(*list(itertools.chain(*[item.values() for item in echo_depth2.values()]))))
necho_depth2 = list(itertools.chain(*list(itertools.chain(*[item.values() for item in necho_depth2.values()]))))
"""
depth = depth.values()
echo_depth = echo_depth.values()
necho_depth = necho_depth.values()
depth2 = depth2.values()
echo_depth2 = echo_depth2.values()
necho_depth2 = necho_depth2.values()
#print(depth)
cdf = CDFPlot()
cdf.set_label('Depth', 'CDF')
cdf.set_data(depth, 'Politics')
cdf.set_data(depth2, 'Other')
cdf.set_legends(['Politics', 'Other'], 'Category')
cdf.save_image('Image/20181002/depth_cdf.png')
cdf = CDFPlot()
cdf.set_label('Depth', 'CDF')
cdf.set_data(echo_depth, 'Echo Chamber')
cdf.set_data(necho_depth, 'Non Echo Chamber')
cdf.set_title('Politics')
cdf.set_legends(['Echo Chamber', 'Non Echo Chamber'], 'User Type')
cdf.save_image('Image/20181002/echo_depth_cdf.png')
cdf = CDFPlot()
cdf.set_label('Depth', 'CDF')
cdf.set_data(echo_depth2, 'Echo Chamber')
cdf.set_data(necho_depth2, 'Non Echo Chamber')
cdf.set_title('Non Politics')
cdf.set_legends(['Echo Chamber', 'Non Echo Chamber'], 'User Type')
cdf.save_image('Image/20181002/echo_depth_cdf2.png')
cdf = CCDFPlot()
cdf.set_label('Depth', 'CCDF')
#cdf.set_log(True)
#cdf.set_ylog()
cdf.set_data(depth)
cdf.set_data(depth2)
cdf.set_legends(['Politics', 'Other'], 'Category')
cdf.save_image('Image/20181002/depth_ccdf.png')
def draw_cdf_plot(datas, datatype, legend, legend_type, filename):
cdf = CDFPlot()
cdf.set_label(datatype, 'CDF')
cdf.set_log(True)
for i in range(len(datas)):
cdf.set_data(datas[i], legend[i])
cdf.set_legends(legend, legend_type)
cdf.save_image('Image/%s/%s.png'%(foldername, filename))
def time_to_depth_echo_chamber(filename):
_, _, time_depth, _, user_depth = get_depth_time_series(None)
print(len(time_depth))
#with open('Data/time_series_data.json', 'w') as f:
# json.dump({'time_depth' : time_depth, 'user_depth' : user_depth}, f)
#with open('Data/time_series_data.json', 'r') as f:
# data = json.load(f)
#time_depth = data['time_depth']
#user_depth = data['user_depth']
print("time series data load done ")
echo_chamber_values = {}
non_echo_chamber_values = {}
for item in ['time_depth', 'user_depth']:
echo_chamber_values[item] = {}
non_echo_chamber_values[item] = {}
for i in range(1,20):
echo_chamber_values[item][i] = []
non_echo_chamber_values[item][i] = []
Bot = bot.load_bot()
echo_chamber_cascade_root = {}
cascade_veracity = {}
echo_chamber_users = e_util.get_echo_chamber_users(filename)
files = os.listdir('RetweetNew')
#collect echo chamber user participate cascade
#for postid in echo_chamber_users.keys():
for postid in files:
v = veracity_type(postid).title()
#get origin tweet of echo chamber user
with open('RetweetNew/%s'%postid, 'r') as f:
tweets = json.load(f)
for tweet in tweets.values():
try:
#if tweet['user'] in echo_chamber_users[postid].keys():
origin = tweet['origin']
otid = tweet['origin_tweet']
#if origin in echo_chamber_users[postid].keys():
if tweet['user'] in echo_chamber_users[postid].keys():
echo_chamber_cascade_root[tweet['origin_tweet']] = 1
except KeyError :
pass
cascade_veracity[tweet['origin_tweet']] = v
print("echo chamber cascade extraction done")
echo_chamber_cascades = echo_chamber_cascade_root.keys()
print('echo chamber cascades')
#print(echo_chamber_cascades)
e = {}; n = {}; r = {}; #echo, non echo, ranked echo
for item in ['True', 'False', 'Mixed']:
e[item] = {}
n[item] = {}
r[item] = {}
for d_type in ['user_depth', 'time_depth']:
e[item][d_type] = {}
n[item][d_type] = {}
r[item][d_type] = {}
for i in range(1, 20):
e[item][d_type][i] = []
n[item][d_type][i] = []
r[item][d_type][i] = []
for key in time_depth.keys():
v = cascade_veracity[key]
if v !='True' and v != 'False':
v = 'Mixed'
if key in echo_chamber_cascades:
#for i in range(1, max(time_depth[key].keys())+1):
for i in range(1, max(time_depth[key].keys())+1):
try:
echo_chamber_values['time_depth'][i].append(time_depth[key][i])
echo_chamber_values['user_depth'][i].append(user_depth[key][i])
e[v]['time_depth'][i].append(time_depth[key][i])
e[v]['user_depth'][i].append(user_depth[key][i])
except KeyError:
pass
else:
for i in range(1, max(time_depth[key].keys())+1):
try :
non_echo_chamber_values['time_depth'][i].append(time_depth[key][i])
non_echo_chamber_values['user_depth'][i].append(user_depth[key][i])
n[v]['time_depth'][i].append(time_depth[key][i])
n[v]['user_depth'][i].append(user_depth[key][i])
except KeyError:
pass
box = BoxPlot(1)
box.set_multiple_data([echo_chamber_values['time_depth'], non_echo_chamber_values['time_depth']])
box.set_ylog()
box.set_label('Depth', 'Minutes to Depth')
box.save_image('%s/time_depth_echo_chamber_box.png'%foldername)
print(echo_chamber_values['time_depth'])
#draw time to depth, user to depth of cascade for echo chamber users participated or non echo chamer users participated
with open('Data/Figure/5_2_1.json', 'w') as f:
json.dump([echo_chamber_values['time_depth'], non_echo_chamber_values['time_depth']], f)
draw_time_to_depth_echo_chamber([echo_chamber_values['time_depth'], non_echo_chamber_values['time_depth']], ['echo chamber', 'no echo chamber'], 'median minutes', 'time_depth_echo_chamber_line')
draw_time_to_depth_echo_chamber([echo_chamber_values['user_depth'], non_echo_chamber_values['user_depth']], ['echo chamber', 'no echo chamber'], 'median unique users', 'user_depth_echo_chamber_line')
with open('Data/Figure/5_2_time.json', 'w') as f:
json.dump({'e':echo_chamber_values['time_depth'][1], 'ne':non_echo_chamber_values['time_depth'][1]}, f)
#draw cdf with top retweet
cdf = CDFPlot()
cdf.set_label('Propagation Time', 'CDF')
cdf.set_log(True)
#cdf.set_ylog()
cdf.set_data(echo_chamber_values['time_depth'][1], '')
cdf.set_data(non_echo_chamber_values['time_depth'][1], '')
cdf.save_image('Image/20181105/depth_propagation_time_cdf.png')
"""
def draw_cdf_plot(datas,
datatype,
legend,
legend_type,
filename,
log_scale=True):
cdf = CDFPlot()
cdf.set_label(datatype, 'CDF')
cdf.set_log(log_scale)
for i in range(len(datas)):
cdf.set_data(datas[i], '')
#ticks = np.arange(-1, 1.1, 0.1)
#ticks = [round(item,1) for item in ticks]
#print(ticks)
#cdf.set_xticks(ticks, index=ticks)
cdf.set_xticks([-1, 0, 1], index=[-1, 0, 1])
#cdf.set_xticks(['0', '1m', '5m', '1h', '1d', '30d', '6m'], index=[0,1,5,60, 24*60, 24*30*60, 24*30*6*60])
if len(legend) > 1:
cdf.set_legends(legend, legend_type)
cdf.save_image(filename)