Yelp Data Graphing

gistfile1.txt

import pandas as pd
import matplotlib.pylab as plt
import datetime
import seaborn as sns
all_data = pd.DataFrame(columns=['Rating', 'Status', 'all', 'Company'])

dates = pd.read_excel('nowaitdates.xlsx')
companies = dates['Company'].tolist()
data = pd.read_excel('new.xlsx')


for company in companies:
    date = dates.loc[dates['Company'] == company]['Date'].iloc[0]
    data['Rating'] = pd.to_numeric(data['Rating'], downcast='float')
    df = data.loc[data['Company'] == company].drop('Company', axis=1)
    df = df.groupby([pd.Grouper(freq='MS',key='Date')]).mean().unstack()
    df = df.reset_index(level=0, drop=True)
    before_NoWait = df.loc[df.index < date]
    after_NoWait = df.loc[df.index > date]
    
    de = pd.DataFrame(df)
    de['k'] = df.index
    de['Status'] = 'Blank'
    de['Status'].loc[de['k'] >= date] = 'After NoWait Implementation'
    de['Status'].loc[de['k'] < date] = 'Before NoWait implementation'
    de.drop(['k'], axis=1, inplace=True)
    de.columns = ['Rating', 'Status']
    de['all'] = ''
    de['Company'] = company
    all_data = pd.concat([all_data, de])

####Seaborn
sns.set_style("whitegrid")
p = sns.boxplot(x = 'Company', y='Rating', hue='Status', data=all_data, palette="Blues", showmeans=True)
p.set_xticklabels(p.get_xticklabels(), rotation=-10)
plt.show()