ใครไปวิ่งงาน CMU Marathon 2019 บ้างจ๊ะ

import pandas as pd
import requests
import seaborn as sns
import matplotlib.pyplot as plt

URL = 'https://go.cmu-marathon.com/retro-2019/runner/tableRunner?init=1&order=asc&offset=0&limit=10000'
data = requests.get(URL).json()
marathon_df = pd.DataFrame(data['rows'])

d = {'ชาย': 'M', 'หญิง': 'F'}
marathon_df['gender'] = marathon_df.gender.map(lambda x: d[x])

def clean_string(s):
    s = s.replace('ฟันรัน', 'Fun Run').replace('ชาย', '')
    s = s.replace('หญิง', '').replace('ชาวต่างชาติ', 'foreigner')
    s = s.replace('ขึ้นไป', '>')
    s = s.replace('ปี', 'yo').replace('ฮาล์ฟมาราธอน', 'Half Marathon')
    s = s.replace('มินิมาราธอน', 'Mini Marathon')
    s = s.replace('มาราธอน', 'Marathon')
    return s

marathon_df['courseList'] = marathon_df.courseList.map(lambda x: clean_string(x))

# เตรียมข้อมูลเพื่อตีกราฟ
sum_df = marathon_df.groupby(['courseList', 'gender']).size().reset_index().rename(columns={0: 'Number of runners'})

order = [
    'Fun Run', 'Mini Marathon  (16-19 yo)', 'Mini Marathon  (20-29 yo)',
    'Mini Marathon  (30-39 yo)', 'Mini Marathon  (40-49 yo)',
    'Mini Marathon  (50-59 yo)', 'Mini Marathon  (60 yo>)', 
    'Half Marathon  (20-29 yo)',
    'Half Marathon  (30-39 yo)', 'Half Marathon  (40-49 yo)',
    'Half Marathon  (50-59 yo)', 'Half Marathon  (60 yo>)',
    'Marathon  (20-29 yo)', 'Marathon  (30-39 yo)',
    'Marathon  (40-49 yo)', 'Marathon  (50-59 yo)',
    'Marathon  (60 yo>)', 'Marathon foreigner ']

# plot จริง
g = sns.catplot(x='courseList', 
                y='Number of runners', 
                hue='gender', 
                kind='bar', 
                data=sum_df, 
                order=order,
                palette="muted",
                size=7)
g.set_xticklabels(rotation=90, fontsize=15)
g.set_yticklabels(fontsize=15)
g.set_ylabels(fontsize=15)
g.set_xlabels(label='', fontsize=15)
g.despine(left=True)

ratios = []
for c, df in marathon_df.groupby('courseList'):
    try:
        ratio = len(df[df.gender == 'F']) / len(df[df.gender == 'M'])
        total = len(df)
    except:
        ratio = 0
    ratios.append({'courseList': c, 'y': ratio, 'x': total})
ratio_df = pd.DataFrame(ratios)
ax = ratio_df.plot(x='x', 
                  y='y', 
                  kind='scatter', 
                  figsize=(6, 6), fontsize=15)
ax.set_xlabel('Number of Runners', fontsize=15)
ax.set_ylabel('Female/Male Ratio', fontsize=15)

ratio_df[['x', 'y', 'courseList']].apply(lambda x: ax.text(*x),axis=1);
plt.axhline(y=1, linewidth=1, color='grey')

size_df = marathon_df.groupby(['Shirt Size', 'gender'])\
    .size().reset_index().rename(columns={0: 'Number of runners'})

g = sns.catplot(x='Shirt Size', 
                y='Number of runners', 
                hue='gender', 
                kind='bar', 
                data=size_df, 
                order=['3S', 'SS', 'S', 'M', 'L', 'XL', '2XL', '3XL', '4XL'],
                palette="muted",
                height=7)
g.set_xticklabels(rotation=90, fontsize=15)
g.set_yticklabels(fontsize=15)
g.set_ylabels(fontsize=15)
g.set_xlabels(label='', fontsize=15)
g.despine(left=True)