Lab on Key Python Lib.ipynb

Kernel: Python 3

Analyze given data set

• Check data shape

• Describe data

• Check for null values and drop null values or replace it

• Drop colums that you find unnecessary

• What is revenue distribution for Bekins in 2005

• What is the growth% for Data/Technology in 2006

• What is the Revenue distribution for Education in 2005

• What is number of companies in Each industry

• Draw a correlation plot to check on feature realtionship.

import numpy as  np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline
sns.set()
from subprocess import check_output

xls = pd.ExcelFile('CompanyDetails.xlsx')
overview = pd.read_excel(xls, 'Overview')
comp = pd.read_excel(xls, 'Financials')

data[data["2005 Revenue"]="3 Round"]]

#xls.shape
overview.head()

comp.head()

df = overview.join(comp, lsuffix='', rsuffix='_right')
df.shape

(529, 18)

df.describe(include="all")

df = df.drop(['ID_right' , 'Name_right'], axis=1)

df.columns

Index(['ID', 'Name', 'Industry', 'Year Founded', 'Employees', 'State', 'City',
       'Zip Code', '2005 Revenue', '2005 Expenses', '2005 Growth%',
       '2004 Revenue', '2004 Expenses', '2004 Growth%', '2003 Revenue',
       '2003 Expenses'],
      dtype='object')

df.head()

df.isnull().sum()

ID                 0
Name               0
Industry           0
Year Founded       1
Employees          0
State              0
City              33
Zip Code          37
2005 Revenue       0
2005 Expenses      0
2005 Growth%       0
2004 Revenue       0
2004 Expenses      0
2004 Growth%      41
2003 Revenue      41
2003 Expenses    394
dtype: int64

df = df.drop(['City' , 'Zip Code' , '2004 Growth%','2003 Revenue','2003 Expenses'], axis=1)

df["Year Founded"]= df["Year Founded"].fillna(df["Year Founded"].median)

df = df.replace("Data/Technology,", "Data/Technology")

df = df.replace("Housing/Real Estate,", "Housing/Real Estate")

df.to_csv('file_name.csv')

df_copy = df

What is the growth% for Data/Technology in 2005

df_copy.groupby(['Industry']).mean()["2005 Growth%"]["Data/Technology"]

2.952745423823052

What is the Revenue distribution for Education in 2005

x = df_copy.loc[df_copy["Industry"] == "Education", "2005 Revenue"]
ax = sns.boxplot(x)

What is number of companies in Each industry

df_copy.groupby(['Industry']).count()["ID"]

Industry
Aerospace and Defense         1
Business & Legal Services    45
Data/Technology              98
Education                    19
Energy                       28
Environment & Weather        12
Finance & Investment         75
Food & Agriculture            6
Geospatial/Mapping           30
Governance                   43
Healthcare                   40
Housing/Real Estate          21
Insurance                    11
Lifestyle & Consumer         25
Media                         1
Research & Consulting        28
Scientific Research          17
Software                      1
Transportation               28
Name: ID, dtype: int64

Draw a correlation plot to check on feature realtionship

ch = df_copy.groupby(['Industry']).mean()
ch

ch.columns

Index(['ID', 'Employees', '2005 Revenue', '2005 Expenses', '2005 Growth%',
       '2004 Revenue', '2004 Expenses'],
      dtype='object')

#df = df.drop(['City' , 'Zip Code' , '2004 Growth%','2003 Revenue','2003 Expenses'], axis=1)

coordata = ch.drop(['ID'],axis=1)

correlations = coordata.corr()

plt.matshow(coordata.corr())
#fig = plt.figure(figsize=(11,9))
plt.xticks(range(len(coordata.columns)), coordata.columns)
plt.yticks(range(len(coordata.columns)), coordata.columns)
plt.colorbar()
plt.show()

# Compute the correlation matrix
corr = coordata.corr()

# Generate a mask for the upper triangle
#mask = np.triu(np.ones_like(corr, dtype=np.bool))

# Set up the matplotlib figure
f, ax = plt.subplots(figsize=(11, 9))

# Generate a custom diverging colormap
cmap = sns.diverging_palette(220, 10, as_cmap=True)

# Draw the heatmap with the mask and correct aspect ratio
sns.heatmap(corr, cmap=cmap, vmax=.3, center=0,annot=True,
            square=True, linewidths=.5, cbar_kws={"shrink": .5})

<matplotlib.axes._subplots.AxesSubplot at 0x21dd9cec288>