Pandas Dataframes & Cufflinks Plots

%%sh
pip install pandas xlrd cufflinks &> /dev/null

#plotly.offline doesn't push your charts to the clouds
import plotly.offline as pyo
#allows us to create the Data and Figure objects
from plotly.graph_objs import *
#plotly.plotly pushes your charts to the cloud  
import plotly.plotly as py
import plotly.figure_factory as ff
from plotly import tools
from plotly.figure_factory import create_table as tbl
import pandas as pd
import numpy as np
import warnings
import cufflinks as cf
cf.go_offline()
cf.set_config_file(theme='white')
warnings.filterwarnings('ignore')

def get_data(fname=Original-tweets_new.xlsx↩, sheet_name='Clean Data', required_fields=['Tweet', 'S/A'], binary=False):
    df = pd.read_excel(fname, sheet_name=sheet_name, encoding='utf-8')
    if binary:
        df = df[(df['S/A'] == 'positive') | (df['S/A'] == 'negative')]
    return df[required_fields]

senti_df = get_data(required_fields=['Tweet', 'Annotation/Type', 'S/A'])

# convert the 'S/A' column to lowercase to eliminate differences between Positive and positive .. etc
senti_df.iloc[:, 1:] = senti_df.iloc[:, 1:].apply(lambda x: x.str.lower())

senti_df.rename(columns={'Annotation/Type': 'Annotation'}, inplace=True)