Baby names iPython notebooks¶
- By David Taylor, www.prooffreader.com
- using data from United States Social Security Administration
- I am making this public to give a head start to those who want to explore this dataset, so they don't have to download and format the data and the python objects used to do preliminary analysis. Please let me know if you find this helpful!
Printing graphs of names that match a list of names¶
List of Mythological names¶
Only Greek, Egyptian, Norse and Roman names are used, because there was too much confusion with others, e.g. Ora was a common girls' name with a Latin origin, but coincidentally was also a figure in Balto-Slavic mythology.
Note: this is an interactive script with repeated code (alas, not yet in functions) that shows the process of getting the data desired, not just the final result.¶
First pass with raw list¶
In [6]:
listed_path = "lists/mythological_names_eg_gk_ro_no.list"
totals_title = "Mythological names in U.S. Social Security baby names database, 1880-2013"
top_cutoff = 6
top_boys_title = "Top %d mythological boys' names from U.S. Social Security database, 1880-2013" % (top_cutoff)
top_girls_title = "Top %d mythological girls' names from U.S. Social Security database, 1880-2013" % (top_cutoff)
last_year = 2013 #change this when Social Security database is updated
save_path = "user_charts" # files created by this notebook will be saved in this directory
import time
import os
if not os.path.isdir(save_path): # creates path if it does not exist
os.makedirs(save_path)
import math
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn # comment out if you don't have it, but it makes good-looking charts
print 'This is standard output from download_and_process.py'
%run download_and_process.py
print '--------------------\nFirst 80 characters of list:'
listed_file = open(listed_path, "r").read()
print listed_file[:80] + ' ...'
all_listed = eval(listed_file) # make sure you trust this file!
all_listed_set = set(all_listed) # to remove duplicates
all_listed = list(all_listed)
print "all_listed: list of length", len(all_listed)
# reduce names dataframe to those matching list
print '--------------------\nDataframe names filtered to those that match list'
print "%d records to begin." % (len(names))
names_listed = names[names.name.isin(all_listed)].copy()
names_listed.sort('pct_max', ascending=False, inplace=True)
print "%d records remaining." % (len(names_listed))
listed_in_df = list(names_listed.name)
print names_listed.head(10)
listed_m = list(names[(names.sex == 'M') & (names.name.isin(listed_in_df))]['name'])
listed_f = list(names[(names.sex == 'F') & (names.name.isin(listed_in_df))]['name'])
#reduce yob dataframe to those matching list
print '--------------------\nDataframe yob filtered to those that match list (count only)'
print "%d records to begin." % (len(yob))
yob_listed = yob[yob.name.isin(listed_in_df)].copy()
yob_listed.sort(['year', 'sex', 'name'], ascending=False, inplace=True)
print "%d records remaining." % (len(yob_listed))
# m and f totals
yob_listed_f_agg = pd.DataFrame(yob_listed[yob_listed.sex == 'F'].groupby('year').sum())[['births', 'pct']]
yob_listed_m_agg = pd.DataFrame(yob_listed[yob_listed.sex == 'M'].groupby('year').sum())[['births', 'pct']]
print '--------------------\nHead of total matching list per year, female'
print yob_listed_f_agg.head()
# print chart of m and f totals
print '\n'
# function to determine a nice y-axis limit a little above the maximum value
# rounds maximum y up to second-most-significant digit
def determine_y_limit(x):
significance = int(math.floor((math.log10(x))))
val = math.floor(x / (10 ** (significance - 1))) + 1
val = val * (10 ** (significance - 1))
return val
#data
xf = list(yob_listed_f_agg.index)
xm = list(yob_listed_m_agg.index)
plt.figure(figsize=(16,9))
plt.plot(xf, list(yob_listed_f_agg.pct), color="red")
plt.plot(xm, list(yob_listed_m_agg.pct), color="blue")
plt.ylim(0, determine_y_limit(max(list(yob_listed_f_agg.pct)
+list(yob_listed_m_agg.pct))))
plt.xlim(1880, 2013)
plt.title(totals_title, fontsize = 20)
plt.xlabel("Year", fontsize = 14)
plt.ylabel("% of total births of that sex", fontsize = 14)
plt.show()
#function to make dataframe for top names
def top_df(yobdf, names, sexes):
""" yobdf = dataframe derived from yob; normally it would just be yob itself.
names = list of names
sexes = list of length 1 for all the same sex, or same length as names. 'F' and 'M' allowed
"""
df_chart = yobdf.copy()
assert len(sexes) == 1 or len(names) == len(sexes)
if len(sexes) == 1:
sexes = sexes * len(names)
df_chart = df_chart[df_chart['name'].isin(names)]
df_chart['temp'] = 0
for row in range(len(df_chart)):
for pos in range(len(names)):
if df_chart.name.iloc[row] == names[pos] and df_chart.sex.iloc[row] == sexes[pos]:
df_chart.temp.iloc[row] = 1
df_chart = df_chart[df_chart.temp == 1]
print "Tail of dataframe:"
print df_chart.tail()
output_df = pd.DataFrame(pd.pivot_table(df_chart, values='pct', index = 'year', columns=['name', 'sex']))
col = output_df.columns[0]
for yr in range(1880, last_year + 1): #inserts missing years
if yr not in output_df.index:
#output_df[col][yr] = 0.0
output_df = output_df.append(pd.DataFrame(index=[yr], columns=[col], data=[0.0]))
output_df = output_df.fillna(0)
return output_df
listed_top_m = top_df(yob, listed_m[:top_cutoff], ['M'])
listed_top_f = top_df(yob, listed_f[:top_cutoff], ['F'])
#a single function to make the four different kinds of charts
def make_chart(df, form='line', title='', colors= [], smoothing=0, \
groupedlist = [], baseline='sym', png_name=''):
dataframe = df.copy()
startyear = min(list(dataframe.index))
endyear = max(list(dataframe.index))
yearstr = '%d-%d' % (startyear, endyear)
legend_size = 0.01
has_male = False
has_female = False
has_both = False
max_y = 0
for name, sex in dataframe.columns:
max_y = max(max_y, dataframe[(name, sex)].max())
final_name = name
if sex == 'M': has_male = True
if sex == 'F': has_female = True
if smoothing > 0:
newvalues = []
for row in range(len(dataframe)):
start = max(0, row - smoothing)
end = min(len(dataframe) - 1, row + smoothing)
newvalues.append(dataframe[(name, sex)].iloc[start:end].mean())
for row in range(len(dataframe)):
dataframe[(name, sex)].iloc[row] = newvalues[row]
if has_male and has_female:
y_text = "% of births of indicated sex"
has_both = True
elif has_male:
y_text = "Percent of male births"
else:
y_text = "Percent of female births"
num_series = len(dataframe.columns)
if colors == []:
colors = ['#BB2114', '#0C5966', '#BA7814', '#4459AB', '#6B3838',
'#B8327B', '#2B947F', '#0D83B5', '#684287', '#8C962C',
'#92289E', '#242D7D']
# my own list of dark contrasting colors
num_colors = len(colors)
if num_series > num_colors:
print "Warning: colors will be repeated."
if title == '':
if num_series == 1:
title = "Popularity of baby name %s in U.S., %s" % (final_name, yearstr)
else:
title = "Popularity of baby names in U.S., %s" % (yearstr)
x_values = range(startyear, endyear + 1)
y_zeroes = [0] * (endyear - startyear)
if form == 'line':
fig, ax = plt.subplots(num=None, figsize=(16, 9), dpi=300, facecolor='w', edgecolor='w')
counter = 0
for name, sex in dataframe.columns:
color = colors[counter % num_colors]
counter += 1
if has_both:
label = "%s (%s)" % (name, sex)
else:
label = name
ax.plot(x_values, dataframe[(name, sex)], label=label, color=color, linewidth = 3)
ax.set_ylim(0,determine_y_limit(max_y))
ax.set_xlim(startyear, endyear)
ax.set_ylabel(y_text, size = 13)
ax.set_title(title, size = 18)
box = ax.get_position()
ax.set_position([box.x0, box.y0 + box.height * legend_size,
box.width, box.height * (1 - legend_size)])
legend_cols = min(5, num_series)
ax.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05), fancybox=True, shadow=True, ncol=legend_cols)
if form == 'subplots_auto':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum alpha: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
current_ymax = dataframe[(name, sex)].max()
tint = 1.0 * current_ymax / determine_y_limit(max_y)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(current_ymax))
axes[counter].set_xlim(startyear, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[0], alpha=tint, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'subplots_same':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum y axis: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(max_y))
axes[counter].set_xlim(startyear, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[1], alpha=1, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'stream':
plt.figure(num=None, figsize=(20,10), dpi=150, facecolor='w', edgecolor='k')
plt.title(title, size=17)
plt.xlim(startyear, endyear)
if has_both:
yaxtext = 'Percent of births of indicated sex (scale: '
elif has_male:
yaxtext = 'Percent of male births (scale: '
else:
yaxtext = 'Percent of female births (scale: '
scale = str(determine_y_limit(max_y)) + ')'
yaxtext += scale
plt.ylabel(yaxtext, size=13)
polys = plt.stackplot(x_values, *[dataframe[(name, sex)] for name, sex in dataframe.columns],
colors=colors, baseline=baseline)
legendProxies = []
for poly in polys:
legendProxies.append(plt.Rectangle((0, 0), 1, 1, fc=poly.get_facecolor()[0]))
namelist = []
for name, sex in dataframe.columns:
if has_both:
namelist.append('%s (%s)' % (name, sex))
else:
namelist.append(name)
plt.legend(legendProxies, namelist, loc=3, ncol=2)
plt.tick_params(\
axis='y',
which='both', # major and minor ticks
left='off',
right='off',
labelleft='off')
plt.show()
if png_name != '':
filename = save_path + "/" + png_name + ".png"
plt.savefig(filename)
plt.close()
# line charts
make_chart(df=listed_top_m,
form='line', # line , subplots_auto , subplots_same , stream
title=top_boys_title,
colors= [],
smoothing=0,
baseline='zero', # zero , sym , wiggle , weighted_wiggle
)
make_chart(df=listed_top_f,
form='line', # line , subplots_auto , subplots_same , stream
title=top_girls_title,
colors= [],
smoothing=0,
baseline='zero', # zero , sym , wiggle , weighted_wiggle
)
names_listed.reset_index(drop = True, inplace = True)
names_listed.head()
names_listed.to_csv('lists/names_matching_mythological_list.csv')
In [7]:
print names_listed.name.unique()
Refine list¶
Some of these names are only coincidentally mythological, e.g. Seth is an Egyptian god's name, but a common Hebrew name, and Doris is a very minor mythological figure, so probably few parents were even aware of the connection (same with Phoebe, Chlore ... Diana is a tougher call, but I think parents are more likely to not have used the name because of a mythological association). Obviously, it's impossible to reading parents' minds with data abstracted like this, so the only choice is to manually curate names that most obviously come from mythology.
In [8]:
cutoffn = 0
# how many names will remain to evaluate after duplicates removed
from collections import OrderedDict
evallistm = OrderedDict()
evallistf = OrderedDict()
# remove names with more common duplicates in other sex
# this happens frequently in ssa db
for name in listed_m:
try:
pctf = names_listed[(names_listed.sex == 'F') &
(names_listed.name == name)].pct_max.iloc[0]
pctm = names_listed[(names_listed.sex == 'M') &
(names_listed.name == name)].pct_max.iloc[0]
except:
pctf = 98
pctm = 99
if (name not in names_listed[names_listed.sex == 'F'].name.unique() or
pctf < pctm):
evallistm[name] = ''
for name in listed_f:
try:
pctf = names_listed[(names_listed.sex == 'F') &
(names_listed.name == name)].pct_max.iloc[0]
pctm = names_listed[(names_listed.sex == 'M') &
(names_listed.name == name)].pct_max.iloc[0]
except:
pctf = 99
pctm = 98
if (name not in names_listed[names_listed.sex == 'M'].name.unique() or
pctm < pctf):
evallistf[name] = ''
if cutoffn > 0:
assert len(evallistm) > cutoffn
assert len(evallistf) > cutoffn
print evallistm[:cutoffn]
print evallistf[:cutoffn]
else:
print 'Length of lists: %d male, %d female\n' % (len(evallistm), len(evallistf))
print evallistm
print ' '
print evallistf
In [9]:
#manually copy and paste the above lists and assign
#'acc' or 'rej' individually to accept or reject
evallistm = OrderedDict([('Sol', 'rej'), ('Seth', 'rej'), ('Griffin', 'rej'), ('Amon', 'rej'),
('Thor', 'acc'), ('Hercules', 'acc'), ('Ladon', 'rej'), ('Odin', 'acc'),
('Hermes', 'acc'), ('Apollo', 'acc'), ('Osiris', 'acc'), ('Min', 'rej'),
('Clete', 'rej'), ('Zeus', 'acc'), ('Phoenix', 'acc'), ('Amen', 'rej'),
('Mars', 'acc'), ('Ares', 'acc'), ('Loki', 'acc'), ('Nike', 'rej'),
('Ran', 'rej'), ('Mercury', 'acc'), ('Tyr', 'acc'), ('Jupiter', 'acc'),
('Kore', 'rej'), ('Ra', 'acc'), ('Anubis', 'acc'), ('Helios', 'acc'),
('Poseidon', 'acc'), ('Makar', 'rej'), ('Pater', 'rej'), ('Amun', 'rej'),
('Fenris', 'acc'), ('Set', 'rej'), ('Demeter', 'rej'), ('Horus', 'acc'),
('Megale', 'rej'), ('Aten', 'acc'), ('Saturn', 'acc')])
evallistf = OrderedDict([('Athena', 'acc'), ('Delia', 'rej'), ('Minerva', 'acc'), ('Doris', 'rej'), ('Phoebe', 'rej'),
('Chloe', 'rej'), ('Diana', 'rej'), ('Flora', 'rej'), ('Sophia', 'rej'),
('Rhea', 'rej'), ('Venus', 'acc'), ('Vesta', 'acc'), ('Luna', 'rej'),
('Thalia', 'acc'), ('Lucina', 'rej'), ('Gerda', 'rej'), ('Eris', 'acc'),
('Artemis', 'acc'), ('Aphrodite', 'acc'), ('Isis', 'acc'), ('Clio', 'acc'),
('Persephone', 'acc'), ('Melaina', 'rej'), ('Shai', 'rej'), ('Andromeda', 'acc'),
('Lamia', 'acc'), ('Sia', 'rej'), ('Hera', 'acc'), ('Nanna', 'rej'), ('Urania', 'acc'),
('Gaia', 'acc'), ('Khloe', 'rej'), ('Chloris', 'rej'), ('Athene', 'acc'),
('Janus', 'rej'), ('Freyja', 'acc'), ('Valkyrie', 'acc'), ('Ourania', 'acc'),
('Juno', 'acc'), ('Vali', 'acc'), ('Holle', 'rej'), ('Cybele', 'acc'),
('Pelagia', 'rej'), ('Anat', 'rej'), ('Soteria', 'rej'), ('Pallas', 'acc'),
('Fortuna', 'rej'), ('Maat', 'acc'), ('Caliope', 'acc'), ('Chimera', 'acc'),
('Deianeira', 'rej'), ('Agathe', 'rej'), ('Lousia', 'rej'), ('Shu', 'rej'),
('Areion', 'rej'), ('Ceres', 'acc'), ('Areia', 'rej'), ('Saturn', 'rej'),
('Lakinia', 'rej'), ('Tyche', 'acc'), ('Khepri', 'acc'), ('Demeter', 'acc'),
('Nike', 'acc')])
# Note, Demeter and Nike taken from males' list and moved to females'; for some reason it spiked in males higher than in females
# Similarly, Saturn moved from females' to males'
# Test that all names have 'acc' or 'rej' values
final_m = []
final_f = []
names_not_validated = []
for item in evallistm:
if evallistm[item] not in ['acc', 'rej']:
names_not_validated.append(item)
elif evallistm[item] == 'acc':
final_m.append(item)
for item in evallistf:
if evallistf[item] not in ['acc', 'rej']:
names_not_validated.append(item)
elif evallistf[item] == 'acc':
final_f.append(item)
final_all = final_m + final_f
if len(names_not_validated) > 0:
print "The following names do not have 'acc' or 'rej' values: ", names_not_validated
raise exception("Names not validated")
print 'Accepted male names:', final_m
print 'Accepted female names:', final_f
print 'Length: %d male, %d female\n' % (len(final_m), len(final_f))
cutmin = min(len(final_m), len(final_f))
final_m = final_m[:cutmin]
final_f = final_f[:cutmin]
print 'After resizing to %d names each:' % (cutmin)
print 'Accepted male names:', final_m
print 'Accepted female names:', final_f
Redo last block because Saturn, Nike and Demeter were assigned to wrong sex¶
In [10]:
from copy import deepcopy
oldm = deepcopy(evallistm)
oldf = deepcopy(evallistf)
cutoffn = 0
# how many names will remain to evaluate after duplicates removed
from collections import OrderedDict
evallistm = OrderedDict()
evallistf = OrderedDict()
# remove names with more common duplicates in other sex
# this happens frequently in ssa db
for name in listed_m:
try:
pctf = names_listed[(names_listed.sex == 'F') &
(names_listed.name == name)].pct_max.iloc[0]
pctm = names_listed[(names_listed.sex == 'M') &
(names_listed.name == name)].pct_max.iloc[0]
except:
pctf = 98
pctm = 99
if (name not in ['Demeter', 'Nike'] and (name not in names_listed[names_listed.sex == 'F'].name.unique() or
pctf < pctm or name == 'Saturn')):
evallistm[name] = ''
for name in listed_f:
try:
pctf = names_listed[(names_listed.sex == 'F') &
(names_listed.name == name)].pct_max.iloc[0]
pctm = names_listed[(names_listed.sex == 'M') &
(names_listed.name == name)].pct_max.iloc[0]
except:
pctf = 99
pctm = 98
if (name != 'Saturn' and (name not in names_listed[names_listed.sex == 'M'].name.unique() or
pctm < pctf or name in ['Demeter', 'Nike'])):
evallistf[name] = ''
for item in evallistm: # copy from above block
try:
evallistm[item] = oldm[item]
except:
pass
for item in evallistf:
try:
evallistf[item] = oldf[item]
except:
pass
if cutoffn > 0:
assert len(evallistm) > cutoffn
assert len(evallistf) > cutoffn
print evallistm[:cutoffn]
print evallistf[:cutoffn]
else:
print 'Length of lists: %d male, %d female\n' % (len(evallistm), len(evallistf))
print evallistm
print ' '
print evallistf
In [11]:
#manually copy and paste the above lists and assign
#'acc' or 'rej' individually to accept or reject
# 72, 29 and 80 character rule (PEP) do not reach this ->|
#########1#########2#########3#########4#########5#########6#########7#2######9X
evallistm = OrderedDict([('Sol', 'rej'), ('Seth', 'rej'), ('Griffin', 'rej'),
('Amon', 'rej'), ('Thor', 'acc'), ('Hercules', 'acc'),
('Ladon', 'rej'), ('Odin', 'acc'), ('Hermes', 'acc'),
('Apollo', 'acc'), ('Osiris', 'acc'), ('Min', 'rej'),
('Clete', 'rej'), ('Zeus', 'acc'), ('Phoenix', 'acc'),
('Amen', 'rej'), ('Mars', 'acc'), ('Ares', 'acc'),
('Loki', 'acc'), ('Ran', 'rej'), ('Mercury', 'acc'),
('Tyr', 'acc'), ('Jupiter', 'acc'), ('Kore', 'rej'),
('Ra', 'acc'), ('Anubis', 'acc'), ('Helios', 'acc'),
('Poseidon', 'acc'), ('Makar', 'rej'),
('Pater', 'rej'), ('Amun', 'rej'), ('Fenris', 'acc'),
('Set', 'rej'), ('Horus', 'acc'), ('Megale', 'rej'),
('Aten', 'acc')])
evallistf = OrderedDict([('Athena', 'acc'), ('Delia', 'rej'), ('Minerva', 'acc'), ('Doris', 'rej'),
('Phoebe', 'rej'), ('Chloe', 'rej'), ('Diana', 'rej'),
('Flora', 'rej'), ('Sophia', 'rej'), ('Rhea', 'rej'),
('Venus', 'acc'), ('Vesta', 'acc'), ('Luna', 'rej'),
('Thalia', 'acc'), ('Lucina', 'rej'), ('Gerda', 'rej'),
('Eris', 'acc'), ('Artemis', 'acc'),
('Aphrodite', 'acc'), ('Isis', 'acc'), ('Clio', 'acc'),
('Persephone', 'acc'), ('Melaina', 'rej'),
('Shai', 'rej'), ('Andromeda', 'acc'),
('Lamia', 'acc'), ('Sia', 'rej'), ('Hera', 'acc'),
('Nanna', 'rej'), ('Urania', 'acc'), ('Gaia', 'acc'),
('Khloe', 'rej'), ('Chloris', 'rej'),
('Athene', 'acc'), ('Nike', 'acc'), ('Janus', 'rej'),
('Freyja', 'acc'), ('Valkyrie', 'acc'),
('Ourania', 'acc'), ('Juno', 'acc'), ('Vali', 'acc'),
('Holle', 'rej'), ('Cybele', 'acc'), ('Pelagia', 'rej'),
('Anat', 'rej'), ('Soteria', 'rej'), ('Pallas', 'acc'),
('Fortuna', 'rej'), ('Maat', 'acc'), ('Caliope', 'acc'),
('Chimera', 'acc'), ('Deianeira', 'rej'),
('Agathe', 'rej'), ('Lousia', 'rej'), ('Shu', 'rej'),
('Areion', 'rej'), ('Ceres', 'acc'), ('Areia', 'rej'),
('Lakinia', 'rej'), ('Tyche', 'acc'), ('Khepri', 'acc')])
# Note, Demeter and Nike taken from males' list and moved to females'; for some reason it spiked in males higher than in females
# Similarly, Saturn moved from females' to males'
# Test that all names have 'acc' or 'rej' values
final_m = []
final_f = []
names_not_validated = []
for item in evallistm:
if evallistm[item] not in ['acc', 'rej']:
names_not_validated.append(item)
elif evallistm[item] == 'acc':
final_m.append(item)
for item in evallistf:
if evallistf[item] not in ['acc', 'rej']:
names_not_validated.append(item)
elif evallistf[item] == 'acc':
final_f.append(item)
if len(names_not_validated) > 0:
print "The following names do not have 'acc' or 'rej' values: ", names_not_validated
raise exception("Names not validated")
print 'Accepted male names:', final_m
print 'Accepted female names:', final_f
print 'Length: %d male, %d female\n' % (len(final_m), len(final_f))
In [12]:
# manually limit to nice round number
nice_round_number = 100 # if too high, there will be no change
final_m = final_m[:nice_round_number]
final_f = final_f[:nice_round_number]
print 'After manually resizing to nice round number of %d names each:' % (nice_round_number)
print 'Accepted male names:', final_m
print 'Accepted female names:', final_f
In [13]:
# BTW, here's those missassigned (it appears) genders:
print names[names.name == 'Saturn']
print names[names.name == 'Demeter']
print names[names.name == 'Nike']
In [14]:
%run download_and_process.py
# reduce names dataframe to those matching list
# print '--------------------\nDataframe names filtered to those that match list'
# print "%d records to begin." % (len(names))
names_listed = names[((names.name.isin(final_m) & (names.sex == 'M')) |
(names.name.isin(final_f) & (names.sex == 'F')) )].copy()
names_listed.sort('pct_max', ascending=False, inplace=True)
# print "%d records remaining." % (len(names_listed))
# listed_in_df = list(names_listed.name)
# print names_listed.head(10)
# listed_m = list(names[(names.sex == 'M') & (names.name.isin(final_m))]['name'])
# listed_f = list(names[(names.sex == 'F') & (names.name.isin(final_f))]['name'])
#reduce yob dataframe to those matching list
print '--------------------\nDataframe yob filtered to those that match list (count only)'
print "%d records to begin." % (len(yob))
yob_listed_m = yob[(yob.name.isin(final_m)) & (yob.sex == 'M')].copy()
yob_listed_m.sort(['year', 'sex', 'name'], ascending=False, inplace=True)
yob_listed_f = yob[(yob.name.isin(final_f)) & (yob.sex == 'F')].copy()
yob_listed_f.sort(['year', 'sex', 'name'], ascending=False, inplace=True)
print "%d records remaining." % (len(yob_listed))
# m and f totals
yob_listed_f_agg = pd.DataFrame(yob_listed_f.groupby('year').sum())[['births', 'pct']]
yob_listed_m_agg = pd.DataFrame(yob_listed_m.groupby('year').sum())[['births', 'pct']]
print '--------------------\nHead of total matching list per year, female'
print yob_listed_f_agg.head()
# print chart of m and f totals
print '\n'
# function to determine a nice y-axis limit a little above the maximum value
# rounds maximum y up to second-most-significant digit
def determine_y_limit(x):
significance = int(math.floor((math.log10(x))))
val = math.floor(x / (10 ** (significance - 1))) + 1
val = val * (10 ** (significance - 1))
return val
#data
xf = list(yob_listed_f_agg.index)
xm = list(yob_listed_m_agg.index)
plt.figure(figsize=(16,9))
plt.plot(xf, list(yob_listed_f_agg.pct), color="red")
plt.plot(xm, list(yob_listed_m_agg.pct), color="blue")
plt.ylim(0, determine_y_limit(max(list(yob_listed_f_agg.pct)
+list(yob_listed_m_agg.pct))))
plt.xlim(1880, 2013)
plt.title('Top 10 mythological names, boy=blue, girl=red', fontsize = 20)
plt.xlabel("Year", fontsize = 14)
plt.ylabel("% of total births of that sex", fontsize = 14)
plt.show()
In [15]:
# all names_listed, so we can see which ones to aggregate
# cutoff of 10 already done
print names_listed[names_listed.sex == 'M'].head(50)
print ''
print names_listed[names_listed.sex == 'F'].head(50)
In [16]:
# just take top 10
nice_round_number = 10 # if too high, there will be no change
final_m = final_m[:nice_round_number]
final_f = final_f[:nice_round_number]
print 'After manually resizing to nice round number of %d names each:' % (nice_round_number)
print 'Accepted male names:', final_m
print 'Accepted female names:', final_f
In [17]:
names = final_m[:10]
sexes = ['M'] # can be length 1 or same length as names
yearstart=1880
yearend=2013
start = time.time()
df_chart = yob.copy()
if len(sexes) == 1:
sexes = sexes * len(names)
df_chart = df_chart[df_chart['name'].isin(names)]
df_chart['temp'] = 0
for row in range(len(df_chart)):
for pos in range(len(names)):
if df_chart.name.iloc[row] == names[pos] and df_chart.sex.iloc[row] == sexes[pos]:
df_chart.temp.iloc[row] = 1
df_chart = df_chart[df_chart.temp == 1]
#To keep more than one data set for charts in memory, change name of chart_1
chart_1 = pd.DataFrame(pd.pivot_table(df_chart, values='pct', index = 'year', columns=['name', 'sex']))
col = chart_1.columns[0]
for yr in range(yearstart, yearend+1): #inserts missing years
if yr not in chart_1.index:
#chart_1[col][yr] = 0.0
chart_1 = chart_1.append(pd.DataFrame(index=[yr], columns=[col], data=[0.0]))
chart_1 = chart_1.fillna(0)
chart_1.sort(inplace=True, ascending=True)
#a single function to make the four different kinds of charts
def make_chart(df=chart_1, form='line', title='', colors= [], smoothing=0, \
groupedlist = [], baseline='sym', png_name=''):
dataframe = df.copy()
startyear = min(list(dataframe.index))
endyear = max(list(dataframe.index))
yearstr = '%d-%d' % (startyear, endyear)
legend_size = 0.01
has_male = False
has_female = False
has_both = False
max_y = 0
for name, sex in dataframe.columns:
max_y = max(max_y, dataframe[(name, sex)].max())
final_name = name
if sex == 'M': has_male = True
if sex == 'F': has_female = True
if smoothing > 0:
newvalues = []
for row in range(len(dataframe)):
start = max(0, row - smoothing)
end = min(len(dataframe) - 1, row + smoothing)
newvalues.append(dataframe[(name, sex)].iloc[start:end].mean())
for row in range(len(dataframe)):
dataframe[(name, sex)].iloc[row] = newvalues[row]
if has_male and has_female:
y_text = "% of births of indicated sex"
has_both = True
elif has_male:
y_text = "Percent of male births"
else:
y_text = "Percent of female births"
num_series = len(dataframe.columns)
if colors == []:
colors = ["#1f78b4","#ae4ec9","#33a02c","#fb9a99","#e31a1c","#a6cee3",
"#fdbf6f","#ff7f00","#cab2d6","#6a3d9a","#ffff99","#b15928"]
#colors = ['#ff0000', '#b00000', '#870000', '#550000', '#e4e400', '#baba00', '#878700', '#545400', '#00ff00', '#00b000', '#008700', '#005500', '#00ffff', '#00b0b0', '#008787', '#005555', '#b0b0ff', '#8484ff', '#4949ff', '#0000ff', '#ff00ff', '#b000b0', '#870087', '#550055', '#e4e4e4', '#bababa', '#878787', '#545454']
from random import shuffle
shuffle(colors)
num_colors = len(colors)
if num_series > num_colors:
print "Warning: colors will be repeated."
if title == '':
if num_series == 1:
title = "Popularity of baby name %s in U.S., %s" % (final_name, yearstr)
else:
title = "Popularity of baby names in U.S., %s" % (yearstr)
x_values = range(startyear, endyear + 1)
y_zeroes = [0] * (endyear - startyear)
if form == 'line':
fig, ax = plt.subplots(num=None, figsize=(16, 9), dpi=300, facecolor='w', edgecolor='w')
counter = 0
for name, sex in dataframe.columns:
color = colors[counter % num_colors]
counter += 1
if has_both:
label = "%s (%s)" % (name, sex)
else:
label = name
ax.plot(x_values, dataframe[(name, sex)], label=label, color=color, linewidth = 3)
ax.set_ylim(0,determine_y_limit(max_y))
ax.set_xlim(1980, endyear)
ax.set_ylabel(y_text, size = 13)
box = ax.get_position()
ax.set_position([box.x0, box.y0 + box.height * legend_size,
box.width, box.height * (1 - legend_size)])
legend_cols = min(5, num_series)
ax.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05), fancybox=True, shadow=True, ncol=legend_cols)
if form == 'subplots_auto':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum alpha: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
current_ymax = dataframe[(name, sex)].max()
tint = 1.0 * current_ymax / determine_y_limit(max_y)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(current_ymax))
axes[counter].set_xlim(startyear, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[0], alpha=tint, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'subplots_same':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum y axis: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(max_y))
axes[counter].set_xlim(startyear, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[1], alpha=1, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'stream':
plt.figure(num=None, figsize=(20,10), dpi=150, facecolor='w', edgecolor='k')
plt.title(title, size=17)
plt.xlim(startyear, endyear)
if has_both:
yaxtext = 'Percent of births of indicated sex (scale: '
elif has_male:
yaxtext = 'Percent of male births (scale: '
else:
yaxtext = 'Percent of female births (scale: '
scale = str(determine_y_limit(max_y)) + ')'
yaxtext += scale
plt.ylabel(yaxtext, size=13)
polys = plt.stackplot(x_values, *[dataframe[(name, sex)] for name, sex in dataframe.columns],
colors=colors, baseline=baseline)
legendProxies = []
for poly in polys:
legendProxies.append(plt.Rectangle((0, 0), 1, 1, fc=poly.get_facecolor()[0]))
namelist = []
for name, sex in dataframe.columns:
if has_both:
namelist.append('%s (%s)' % (name, sex))
else:
namelist.append(name)
plt.legend(legendProxies, namelist, loc=3, ncol=2)
plt.tick_params(\
axis='y',
which='both', # major and minor ticks
left='off',
right='off',
labelleft='off')
plt.show()
if png_name != '':
filename = save_path + "/" + png_name + ".png"
plt.savefig(filename)
plt.close()
#line graph
make_chart(df=chart_1,
form='stream', # line , subplots_auto , subplots_same , stream
title='',
colors= [],
smoothing=0,
baseline='sym', # zero , sym , wiggle , weighted_wiggle
png_name = '', # if '', will not be saved
)
In [18]:
names = final_f[:12]
sexes = ['F'] # can be length 1 or same length as names
yearstart=1880
yearend=2013
start = time.time()
df_chart = yob.copy()
if len(sexes) == 1:
sexes = sexes * len(names)
df_chart = df_chart[df_chart['name'].isin(names)]
df_chart['temp'] = 0
for row in range(len(df_chart)):
for pos in range(len(names)):
if df_chart.name.iloc[row] == names[pos] and df_chart.sex.iloc[row] == sexes[pos]:
df_chart.temp.iloc[row] = 1
df_chart = df_chart[df_chart.temp == 1]
#To keep more than one data set for charts in memory, change name of chart_1
chart_1 = pd.DataFrame(pd.pivot_table(df_chart, values='pct', index = 'year', columns=['name', 'sex']))
col = chart_1.columns[0]
for yr in range(yearstart, yearend+1): #inserts missing years
if yr not in chart_1.index:
#chart_1[col][yr] = 0.0
chart_1 = chart_1.append(pd.DataFrame(index=[yr], columns=[col], data=[0.0]))
chart_1 = chart_1.fillna(0)
chart_1.sort(inplace=True, ascending=True)
#a single function to make the four different kinds of charts
def make_chart(df=chart_1, form='line', title='', colors= [], smoothing=0, \
groupedlist = [], baseline='sym', png_name=''):
dataframe = df.copy()
startyear = min(list(dataframe.index))
endyear = max(list(dataframe.index))
yearstr = '%d-%d' % (startyear, endyear)
legend_size = 0.01
has_male = False
has_female = False
has_both = False
max_y = 0
for name, sex in dataframe.columns:
max_y = max(max_y, dataframe[(name, sex)].max())
final_name = name
if sex == 'M': has_male = True
if sex == 'F': has_female = True
if smoothing > 0:
newvalues = []
for row in range(len(dataframe)):
start = max(0, row - smoothing)
end = min(len(dataframe) - 1, row + smoothing)
newvalues.append(dataframe[(name, sex)].iloc[start:end].mean())
for row in range(len(dataframe)):
dataframe[(name, sex)].iloc[row] = newvalues[row]
if has_male and has_female:
y_text = "% of births of indicated sex"
has_both = True
elif has_male:
y_text = "Percent of male births"
else:
y_text = "Percent of female births"
num_series = len(dataframe.columns)
if colors == []:
colors = ["#1f78b4","#ae4ec9","#33a02c","#fb9a99","#e31a1c","#a6cee3",
"#fdbf6f","#ff7f00","#cab2d6","#6a3d9a","#ffff99","#b15928"]
#colors = ['#ff0000', '#b00000', '#870000', '#550000', '#e4e400', '#baba00', '#878700', '#545400', '#00ff00', '#00b000', '#008700', '#005500', '#00ffff', '#00b0b0', '#008787', '#005555', '#b0b0ff', '#8484ff', '#4949ff', '#0000ff', '#ff00ff', '#b000b0', '#870087', '#550055', '#e4e4e4', '#bababa', '#878787', '#545454']
from random import shuffle
shuffle(colors)
num_colors = len(colors)
if num_series > num_colors:
print "Warning: colors will be repeated."
if title == '':
if num_series == 1:
title = "Popularity of baby name %s in U.S., %s" % (final_name, yearstr)
else:
title = "Popularity of baby names in U.S., %s" % (yearstr)
x_values = range(startyear, endyear + 1)
y_zeroes = [0] * (endyear - startyear)
if form == 'line':
fig, ax = plt.subplots(num=None, figsize=(16, 9), dpi=300, facecolor='w', edgecolor='w')
counter = 0
for name, sex in dataframe.columns:
color = colors[counter % num_colors]
counter += 1
if has_both:
label = "%s (%s)" % (name, sex)
else:
label = name
ax.plot(x_values, dataframe[(name, sex)], label=label, color=color, linewidth = 3)
ax.set_ylim(0,determine_y_limit(max_y))
ax.set_xlim(1980, endyear)
ax.set_ylabel(y_text, size = 13)
box = ax.get_position()
ax.set_position([box.x0, box.y0 + box.height * legend_size,
box.width, box.height * (1 - legend_size)])
legend_cols = min(5, num_series)
ax.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05), fancybox=True, shadow=True, ncol=legend_cols)
if form == 'subplots_auto':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum alpha: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
current_ymax = dataframe[(name, sex)].max()
tint = 1.0 * current_ymax / determine_y_limit(max_y)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(current_ymax))
axes[counter].set_xlim(startyear, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[0], alpha=tint, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'subplots_same':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum y axis: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(max_y))
axes[counter].set_xlim(startyear, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[1], alpha=1, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'stream':
plt.figure(num=None, figsize=(20,10), dpi=150, facecolor='w', edgecolor='k')
plt.title(title, size=17)
plt.xlim(startyear, endyear)
if has_both:
yaxtext = 'Percent of births of indicated sex (scale: '
elif has_male:
yaxtext = 'Percent of male births (scale: '
else:
yaxtext = 'Percent of female births (scale: '
scale = str(determine_y_limit(max_y)) + ')'
yaxtext += scale
plt.ylabel(yaxtext, size=13)
polys = plt.stackplot(x_values, *[dataframe[(name, sex)] for name, sex in dataframe.columns],
colors=colors, baseline=baseline)
legendProxies = []
for poly in polys:
legendProxies.append(plt.Rectangle((0, 0), 1, 1, fc=poly.get_facecolor()[0]))
namelist = []
for name, sex in dataframe.columns:
if has_both:
namelist.append('%s (%s)' % (name, sex))
else:
namelist.append(name)
plt.legend(legendProxies, namelist, loc=3, ncol=2)
plt.tick_params(\
axis='y',
which='both', # major and minor ticks
left='off',
right='off',
labelleft='off')
plt.show()
if png_name != '':
filename = save_path + "/" + png_name + ".png"
plt.savefig(filename)
plt.close()
#line graph
make_chart(df=chart_1,
form='stream', # line , subplots_auto , subplots_same , stream
title="10 most popular mythological girls' names, 2914-2013",
colors= [],
smoothing=0,
baseline='sym', # zero , sym , wiggle , weighted_wiggle
png_name = '', # if '', will not be saved
)
Values before WWII are HEAVILY waited towards white middle-class, so refine these graphs to show 1945-present¶
In [19]:
names = final_f[:10]
sexes = ['F'] # can be length 1 or same length as names
yearstart=1880 # for data, not graph
yearend=2013
xmin = 1940
start = time.time()
df_chart = yob.copy()
if len(sexes) == 1:
sexes = sexes * len(names)
df_chart = df_chart[df_chart['name'].isin(names)]
df_chart['temp'] = 0
for row in range(len(df_chart)):
for pos in range(len(names)):
if df_chart.name.iloc[row] == names[pos] and df_chart.sex.iloc[row] == sexes[pos]:
df_chart.temp.iloc[row] = 1
df_chart = df_chart[df_chart.temp == 1]
#To keep more than one data set for charts in memory, change name of chart_1
chart_1 = pd.DataFrame(pd.pivot_table(df_chart, values='pct', index = 'year', columns=['name', 'sex']))
col = chart_1.columns[0]
for yr in range(yearstart, yearend+1): #inserts missing years
if yr not in chart_1.index:
#chart_1[col][yr] = 0.0
chart_1 = chart_1.append(pd.DataFrame(index=[yr], columns=[col], data=[0.0]))
chart_1 = chart_1.fillna(0)
chart_1.sort(inplace=True, ascending=True)
#a single function to make the four different kinds of charts
def make_chart(df=chart_1, form='line', title='', colors= [], smoothing=0, \
groupedlist = [], baseline='sym', png_name=''):
dataframe = df.copy()
startyear = min(list(dataframe.index))
endyear = max(list(dataframe.index))
yearstr = '%d-%d' % (startyear, endyear)
legend_size = 0.01
has_male = False
has_female = False
has_both = False
max_y = 0
for name, sex in dataframe.columns:
max_y = max(max_y, dataframe[(name, sex)].max())
final_name = name
if sex == 'M': has_male = True
if sex == 'F': has_female = True
if smoothing > 0:
newvalues = []
for row in range(len(dataframe)):
start = max(0, row - smoothing)
end = min(len(dataframe) - 1, row + smoothing)
newvalues.append(dataframe[(name, sex)].iloc[start:end].mean())
for row in range(len(dataframe)):
dataframe[(name, sex)].iloc[row] = newvalues[row]
if has_male and has_female:
y_text = "% of births of indicated sex"
has_both = True
elif has_male:
y_text = "Percent of male births"
else:
y_text = "Percent of female births"
num_series = len(dataframe.columns)
if colors == []:
colors = ["#1f78b4","#ae4ec9","#33a02c","#fb9a99","#e31a1c","#a6cee3",
"#fdbf6f","#ff7f00","#cab2d6","#6a3d9a","#ffff99","#b15928"]
#colors = ['#ff0000', '#b00000', '#870000', '#550000', '#e4e400', '#baba00', '#878700', '#545400', '#00ff00', '#00b000', '#008700', '#005500', '#00ffff', '#00b0b0', '#008787', '#005555', '#b0b0ff', '#8484ff', '#4949ff', '#0000ff', '#ff00ff', '#b000b0', '#870087', '#550055', '#e4e4e4', '#bababa', '#878787', '#545454']
from random import shuffle
shuffle(colors)
num_colors = len(colors)
if num_series > num_colors:
print "Warning: colors will be repeated."
if title == '':
if num_series == 1:
title = "Popularity of baby name %s in U.S., %s" % (final_name, yearstr)
else:
title = "Popularity of baby names in U.S., %s" % (yearstr)
x_values = range(startyear, endyear + 1)
y_zeroes = [0] * (endyear - startyear)
if form == 'line':
fig, ax = plt.subplots(num=None, figsize=(16, 9), dpi=300, facecolor='w', edgecolor='w')
counter = 0
for name, sex in dataframe.columns:
color = colors[counter % num_colors]
counter += 1
if has_both:
label = "%s (%s)" % (name, sex)
else:
label = name
ax.plot(x_values, dataframe[(name, sex)], label=label, color=color, linewidth = 3)
ax.set_ylim(0,determine_y_limit(max_y))
ax.set_xlim(xmin, endyear)
ax.set_ylabel(y_text, size = 13)
box = ax.get_position()
ax.set_position([box.x0, box.y0 + box.height * legend_size,
box.width, box.height * (1 - legend_size)])
legend_cols = min(5, num_series)
ax.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05), fancybox=True, shadow=True, ncol=legend_cols)
if form == 'subplots_auto':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum alpha: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
current_ymax = dataframe[(name, sex)].max()
tint = 1.0 * current_ymax / determine_y_limit(max_y)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(current_ymax))
axes[counter].set_xlim(xmin, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[0], alpha=tint, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'subplots_same':
counter = 0
fig, axes = plt.subplots(num_series, 1, figsize=(12, 3.5*num_series))
print 'Maximum y axis: %d percent' % (determine_y_limit(max_y))
for name, sex in dataframe.columns:
if sex=='M':
sex_label = 'male'
else:
sex_label = 'female'
label = "Percent of %s births for %s" % (sex_label, name)
axes[counter].plot(x_values, dataframe[(name, sex)], color='k')
axes[counter].set_ylim(0,determine_y_limit(max_y))
axes[counter].set_xlim(xmin, endyear)
axes[counter].fill_between(x_values, dataframe[(name, sex)], color=colors[1], alpha=1, interpolate=True)
axes[counter].set_ylabel(label, size=11)
plt.subplots_adjust(hspace=0.1)
counter += 1
if form == 'stream':
plt.figure(num=None, figsize=(20,16.67), dpi=150, facecolor='w', edgecolor='k')
plt.title(title, size=17)
plt.xlim(xmin, endyear)
if has_both:
yaxtext = 'Percent of births of indicated sex (scale: '
elif has_male:
yaxtext = 'Percent of male births (scale: '
else:
yaxtext = 'Percent of female births (scale: '
scale = str(determine_y_limit(max_y)) + ')'
yaxtext += scale
plt.ylabel(yaxtext, size=13)
polys = plt.stackplot(x_values, *[dataframe[(name, sex)] for name, sex in dataframe.columns],
colors=colors, baseline=baseline)
legendProxies = []
for poly in polys:
legendProxies.append(plt.Rectangle((0, 0), 1, 1, fc=poly.get_facecolor()[0]))
namelist = []
for name, sex in dataframe.columns:
if has_both:
namelist.append('%s (%s)' % (name, sex))
else:
namelist.append(name)
plt.legend(legendProxies, namelist, loc=3, ncol=2)
plt.tick_params(\
axis='y',
which='both', # major and minor ticks
left='off',
right='off',
labelleft='off')
plt.show()
if png_name != '':
filename = save_path + "/" + png_name + ".png"
plt.savefig(filename)
plt.close()
#stream graph
make_chart(df=chart_1,
form='stream', # line , subplots_auto , subplots_same , stream
title='',
colors= [],
smoothing=0,
baseline='sym', # zero , sym , wiggle , weighted_wiggle
png_name = '', # if '', will not be saved
)