import numpy as np
import pandas as pd
from pandas import DataFrame
from pandas import Series

#Set PANDAS to show all columns in DataFrame
pd.set_option('display.max_columns', None)

import matplotlib.pyplot as plt
print matplotlib.__version__

#NECESSARY FOR XTICKS OPTION, ETC.
from pylab import*

%matplotlib inline  

import seaborn as sns
print sns.__version__

df = pd.read_pickle('CSR tweets - 2013 by 41 accounts.pkl')
print len(df)
df.head(2)

df.columns

len(df.columns)

pd.unique(df.from_user_screen_name.ravel())

len(pd.unique(df.from_user_screen_name.ravel()))

len(pd.unique(df.Company.ravel()))

def f(x):
     return Series(dict(Number_of_tweets = x['content'].count(), 
                        Company=x['Company'].min(),
                        Description=x['from_user_description'].min(),
                        ))

account_count = df.groupby('from_user_screen_name').apply(f) 
print len(account_count)
account_count

counts = account_count.groupby('Company').size()  #create a series of number of rows per company
df2 = pd.DataFrame(counts, columns = ['size'])    #convert series to a dataframe
df2 = df2[df2['size']>1]                           #keep only those cases with more than one row in account-level dataframe
df2                                               #show dataframe

account_count.to_csv('Number of Tweets per Account.csv')

DataFrame.sort?

account_count = account_count.sort(['Number_of_tweets'], ascending=False)
account_count

plt.rcParams['figure.figsize'] = (20, 10)

account_count['Number_of_tweets'].plot(kind='bar')

account_count['Number_of_tweets'].plot(kind='barh')

account_plot = account_count['Number_of_tweets'].plot(kind='bar', alpha=0.75)
xticks(fontsize = 10,rotation=60, ha ="right")
account_plot.set_xlabel('Twitter Account Name', weight='bold', labelpad=30)   
account_plot.set_ylabel('# Tweets', weight='bold', labelpad=25) #ADD PADDING TO RIGHT OF Y-AXIS LABEL
savefig('account counts - bar graph.png', bbox_inches='tight', dpi=300, format='png')

account_plot = account_count['Number_of_tweets'].plot(kind='barh', alpha=0.75)
xticks(fontsize = 12) #,rotation=60, ha ="right")
account_plot.set_ylabel('Twitter Account Name', weight='bold', labelpad=20)   
account_plot.set_xlabel('# Tweets', weight='bold', labelpad=25) #ADD PADDING TO RIGHT OF Y-AXIS LABEL
savefig('account counts - horizontal bar graph.png', bbox_inches='tight', dpi=300, format='png')