#dataset downloaded from ourworldindata.org import os import matplotlib.pyplot as plt %matplotlib inline import seaborn as sns import numpy as np import pandas as pd

df = pd.read_csv('Total Oppurtunity.csv') df.head()

df.info()

import datetime as dt df['Close Date'] = pd.to_datetime(df['Close Date'])

df.info()

df.isnull().sum()

df.dropna(inplace=True)

df.isnull().sum()

df.describe()

df.sample(10)

df.drop(['Closed','# of Losses','# of Open', '# of Wins','Sales', 'Won', 'Group Opportunities'], axis = 'columns')

#df['Account Name'].unique() df.value_counts()

df_final = df.drop(['Closed','# of Losses','# of Open', '# of Wins','Sales', 'Won', 'Group Opportunities'], axis = 'columns')

df_final

df.to_excel('Total Oppurtunity Cleaned.xlsx')