HR Analytic Using Logistic Regression

✏Contents of notebook :-

1. Importing Libraries
2. Exploratory Data Analysis
3. Basic Data Cleaning
4. Data Visulaization
5. Data Preprocessing
6. Model Building

Importing Libraries

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
%matplotlib inline
import warnings
warnings.filterwarnings('ignore')

Importing Dataset

hr = pd.read_csv('HR_comma_sep.csv')

Performing Some EDA

hr.head()

satisfaction_level last_evaluation number_project average_montly_hours time_spend_company Work_accident left promotion_last_5years Depart

0 0.38 0.53 2 157 3 0 1 0

1 0.80 0.86 5 262 6 0 1 0

2 0.11 0.88 7 272 4 0 1 0

3 0.72 0.87 5 223 5 0 1 0

4 0.37 0.52 2 159 3 0 1 0

hr.size

149990

hr.describe()

satisfaction_level last_evaluation number_project average_montly_hours time_spend_company Work_accident left promotion_last_5

count 14999.000000 14999.000000 14999.000000 14999.000000 14999.000000 14999.000000 14999.000000 14999.0

mean 0.612834 0.716102 3.803054 201.050337 3.498233 0.144610 0.238083

std 0.248631 0.171169 1.232592 49.943099 1.460136 0.351719 0.425924

min 0.090000 0.360000 2.000000 96.000000 2.000000 0.000000 0.000000

25% 0.440000 0.560000 3.000000 156.000000 3.000000 0.000000 0.000000

50% 0.640000 0.720000 4.000000 200.000000 3.000000 0.000000 0.000000

75% 0.820000 0.870000 5.000000 245.000000 4.000000 0.000000 0.000000

max 1.000000 1.000000 7.000000 310.000000 10.000000 1.000000 1.000000

 hr.describe(include = 'object')

Department salary

count 14999 14999

unique 10 3

top sales low

freq 4140 7316

hr.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 14999 entries, 0 to 14998
Data columns (total 10 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 satisfaction_level 14999 non-null float64
1 last_evaluation 14999 non-null float64
2 number_project 14999 non-null int64
3 average_montly_hours 14999 non-null int64
4 time_spend_company 14999 non-null int64
5 Work_accident 14999 non-null int64
6 left 14999 non-null int64
7 promotion_last_5years 14999 non-null int64
8 Department 14999 non-null object
9 salary 14999 non-null object
dtypes: float64(2), int64(6), object(2)
memory usage: 1.1+ MB

There is no null values in this dataset so We don't have to perform any data cleaning for null values

hr['Department'].unique()

array(['sales', 'accounting', 'hr', 'technical', 'support', 'management',

'IT', 'product_mng', 'marketing', 'RandD'], dtype=object)

hr['salary'].unique()

array(['low', 'medium', 'high'], dtype=object)

Department

hr['Department'].value_counts()/len(hr)*100

sales 27.601840
technical 18.134542
support 14.860991
IT 8.180545
product_mng 6.013734
marketing 5.720381
RandD 5.247016
accounting 5.113674
hr 4.926995
management 4.200280
Name: Department, dtype: float64

Data Visualtization

fig, ax = plt.subplots(1,2, figsize = (16,8))

sns.set(style = 'dark' , color_codes = True)
data = hr['Department'].value_counts()
pal = sns.color_palette("magma" , len(data))

ax[0] = sns.barplot( x = data.index, y = data.values , ax = ax[0] , palette = pal)

for bar in ax[0].patches:
ax[0].annotate( "{:.0f}".format(bar.get_height()) , ( bar.get_x() + bar.get_width()/2 , bar.get_height()) , ha

ax[0].set_xticklabels( ax[0].get_xticklabels() , rotation = 70)

_,_, autotexts = ax[1].pie( data.values, labels = data.index , autopct = "%.2f%%" , colors = pal)

for text in autotexts:

text.set_color('white')

plt.title("Department")

plt.show()

hr['salary'].value_counts()/len(hr)*100

low 48.776585
medium 42.976198
high 8.247216
Name: salary, dtype: float64

fig, ax = plt.subplots(1,2, figsize = (16,8))

sns.set(style = 'dark' , color_codes = True)
data = hr['salary'].value_counts()
pal = sns.color_palette("magma" , len(data))

ax[0] = sns.barplot( x = data.index, y = data.values , ax = ax[0] , palette = pal)

for bar in ax[0].patches:
ax[0].annotate( "{:.0f}".format(bar.get_height()) , ( bar.get_x() + bar.get_width()/2 , bar.get_height()) , ha

ax[0].set_xticklabels( ax[0].get_xticklabels() , rotation = 70)

_,_, autotexts = ax[1].pie( data.values, labels = data.index , autopct = "%.2f%%" , colors = pal)

for text in autotexts:

text.set_color('white')

plt.title("Salary")

plt.show()
hr.left.value_counts()/len(hr)*100

0 76.191746
1 23.808254
Name: left, dtype: float64

hr['left'] = hr.left.astype('object')

fig, ax = plt.subplots(1,2, figsize = (16,8))

sns.set(style = 'dark' , color_codes = True)
data = hr['left'].value_counts()
pal = sns.color_palette("magma" , len(data))

ax[0] = sns.barplot( x = data.index, y = data.values , ax = ax[0] , palette = pal[::-1])

for bar in ax[0].patches:
ax[0].annotate( "{:.0f}".format(bar.get_height()) , ( bar.get_x() + bar.get_width()/2 , bar.get_height()) , ha

ax[0].set_xticklabels( ax[0].get_xticklabels() , rotation = 70)

_,_, autotexts = ax[1].pie( data.values, labels = data.index , autopct = "%.2f%%" , colors = pal[::-1])

for text in autotexts:

text.set_color('white')

plt.title("Leave")

plt.show()
 ct = pd.crosstab(hr['Department'] , hr['salary'])

ct.div(ct.sum(1).astype(float) , axis = 0).plot(kind = 'bar' , stacked = True)

plt.title('Salary vs Department')

plt.xlabel('Department')

plt.ylabel('Salary')

plt.show()

Now find the independent features

fig, ax = plt.subplots(1,2, figsize = (16,8))

sns.set(style = 'dark' , color_codes = True)
data = hr['left'].value_counts()
pal = sns.color_palette("magma" , len(data))

ax[0] = sns.barplot( x = data.index, y = data.values , ax = ax[0] , palette = pal)

for bar in ax[0].patches:
ax[0].annotate( "{:.0f}".format(bar.get_height()) , ( bar.get_x() + bar.get_width()/2 , bar.get_height()) , ha

ax[0].set_xticklabels( ax[0].get_xticklabels() , rotation = 70)

_,_, autotexts = ax[1].pie( data.values, labels = data.index , autopct = "%.2f%%" , colors = pal)

for text in autotexts:

text.set_color('white')

plt.title("Left")

plt.show()
We can see that almost 24% employees leave the company

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.countplot('Department' , data = hr , palette = sns.color_palette('magma'), hue = 'left')

<AxesSubplot:xlabel='Department', ylabel='count'>

We can see here that there is no such major impact of department on retention of any employee

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.countplot('salary' , data = hr , palette = sns.color_palette('magma'), hue = 'left')

<AxesSubplot:xlabel='salary', ylabel='count'>
We can clearly see here that employees with higher salaries are not like to leave the company

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.boxplot( x = 'left' , y = 'satisfaction_level' , data = hr , palette = sns.color_palette('magma'), hue = 'left'

<AxesSubplot:xlabel='left', ylabel='satisfaction_level'>

We can see here that satisfaction level is directly impact the leaving chances of the employee

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.boxplot( x = 'left' , y = 'last_evaluation' , data = hr , palette = sns.color_palette('magma'), hue = 'left')

<AxesSubplot:xlabel='left', ylabel='last_evaluation'>
From above chart there seem to be no impact of last_evalution on employee retention

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.boxplot( x = 'left' , y = 'number_project' , data = hr , palette = sns.color_palette('magma'), hue = 'left')

<AxesSubplot:xlabel='left', ylabel='number_project'>

From above chart there seem to be no impact of number_project on employee retention

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.boxplot( x = 'left' , y = 'average_montly_hours' , data = hr , palette = sns.color_palette('magma'), hue = 'left'

<AxesSubplot:xlabel='left', ylabel='average_montly_hours'>
From above chart there seem to be some impact of average_montly_hours on employee retention
but it is not too major but we will consider it in our analysis

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.boxplot( x = 'left' , y = 'time_spend_company' , data = hr , palette = sns.color_palette('magma'), hue = 'left'

<AxesSubplot:xlabel='left', ylabel='time_spend_company'>

Above bar chart shows employees with low time_spend_compmay are likely to not leave the
company

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.countplot('Work_accident' , data = hr , palette = sns.color_palette('magma'), hue = 'left')

<AxesSubplot:xlabel='Work_accident', ylabel='count'>
From above chart there seem to be impact of Work_accident on employee retention

fig = plt.figure( figsize = (12,8))

sns.set(style = 'dark' , color_codes = True)
sns.countplot('promotion_last_5years' , data = hr , palette = sns.color_palette('magma'), hue = 'left')

<AxesSubplot:xlabel='promotion_last_5years', ylabel='count'>

From the data analysis so far we can conclude that we will use following variables as independant
variables in our model
1. **Satisfaction Level**
2. **Average Monthly Hours**
3. **Promotion Last 5 Years**
4. **Salary**
5. **Work Accident**

Data Preprocessing

subdf = hr[['satisfaction_level','average_montly_hours','promotion_last_5years','Work_accident','salary']]
subdf.head()

satisfaction_level average_montly_hours promotion_last_5years Work_accident salary

0 0.38 157 0 0 low

1 0.80 262 0 0 medium

2 0.11 272 0 0 medium

3 0.72 223 0 0 low

4 0.37 159 0 0 low

salary_dummies = pd.get_dummies(subdf.salary, prefix="salary")

df_with_dummies = pd.concat([subdf,salary_dummies],axis='columns')

df_with_dummies.head()

satisfaction_level average_montly_hours promotion_last_5years Work_accident salary salary_high salary_low salary_medium

0 0.38 157 0 0 low 0 1 0

1 0.80 262 0 0 medium 0 0 1

2 0.11 272 0 0 medium 0 0 1

3 0.72 223 0 0 low 0 1 0

4 0.37 159 0 0 low 0 1 0

df_with_dummies.drop(['salary','salary_low'],axis='columns',inplace=True)
df_with_dummies.head()

satisfaction_level average_montly_hours promotion_last_5years Work_accident salary_high salary_medium

0 0.38 157 0 0 0 0

1 0.80 262 0 0 0 1

2 0.11 272 0 0 0 1

3 0.72 223 0 0 0 0

4 0.37 159 0 0 0 0

X = df_with_dummies
X.head()

satisfaction_level average_montly_hours promotion_last_5years Work_accident salary_high salary_medium

0 0.38 157 0 0 0 0

1 0.80 262 0 0 0 1

2 0.11 272 0 0 0 1

3 0.72 223 0 0 0 0

4 0.37 159 0 0 0 0

y = hr['left'].astype(str)
y

0 1
1 1
2 1
3 1
4 1
..
14994 1
14995 1
14996 1
14997 1
14998 1
Name: left, Length: 14999, dtype: object

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.3)

Model Building

from sklearn.linear_model import LogisticRegression

 model = LogisticRegression()

model.fit(X_train, y_train)

LogisticRegression()

ypred = model.predict(X_test)

model.score(X_test,y_test)

0.7753333333333333

Exciting Milestone: Successfully trained my first logistic regression model, one more step in my
journey into data science and predictive analytics. Looking forward to exploring more complex
algorithms and applications!
Loading [MathJax]/extensions/Safe.js