Data exploration sets and developing a deep understanding of the data is one of the most important skills every data scientist should possess. People sometimes estimate that the time spent on these activities can go as high as 80% of the project time. Python has been gaining much ground as a preferred tool for data scientists lately, and for the right reasons. Ease of learning, powerful libraries with integration of C/C++, production readiness, and integration with the web stack are some of the main reasons for this move lately.

I will use NumPy, Matplotlib, Seaborn, and Pandas to explore data in this guide. These are powerful libraries that can perform data exploration in Python. The idea is to create a ready reference for some of the regular operations that are required frequently. I am using an iPython Notebook to perform data exploration and would recommend the same for its natural fit for exploratory analysis.

In case you missed it, I suggest you refer to the Python baby steps series to understand the basics of programming.

- Learning Python for data analysis – with instructions on installation and creating the environment
- Libraries and data structures
- Exploratory analysis in Python (using Pandas)
- Data Munging in Python (using Pandas)

- Introduction
- Data Exploration in Python using NumPy, Matplotlib and Pandas
- How do I load data file(s) using Pandas?
- How to convert a variable to a different data type?
- How to transpose a Data set or dataframe using Pandas?
- How to sort a Pandas DataFrame?
- How to create plots (Histogram, Scatter, Box Plot)?
- How to generate frequency tables with Pandas?
- How to do sample Data set in Python?
- How do duplicate values of a variable in a Pandas Dataframe be removed?
- How to group variables in Pandas to calculate count, average, sum?
- How to recognize and Treat missing values and outliers in Pandas?
- How to merge / join data sets and Pandas dataframes?
- Conclusion

- Frequently Asked Questions

Input data sets can be in various formats (XLS, TXT, CSV, JSON ). Python makes it easy to load data from any source due to its simple syntax and availability of predefined libraries, such as Pandas. Here, I will use Pandas itself.

Pandas features many functions for reading tabular data as a Pandas DataFrame object. Below are the common functions that can be used to read data (including read_csv in Pandas):

**Code**

```
mport pandas as pd
#Import Library Pandas
df = pd.read_csv("train.csv") #I am working in Windows environment
#Reading the dataset in a dataframe using Pandas
print(df.head(3)) #Print first three observations
```

Code

df=pd.read_excel("E:/EMP.xlsx", "Data") # Load Data sheet of excel file EMPOutput

print df

**Code:**

df=pd.read_csv("E:/Test.txt",sep='\t') # Load Data from text file having tab '\t' delimeter print df

Converting a variable data type to others is an important and common procedure after loading data. Let’s look at some of the commands to perform these conversions:

srting_outcome = str(numeric_input) #Converts numeric_input to string_outcome integer_outcome = int(string_input) #Converts string_input to integer_outcome float_outcome = float(string_input) #Converts string_input to integer_outcome

The later operations are especially useful when you input value from user using raw_input(). By default, the values are read at string.

There are multiple ways to do this. The simplest would be to use the *datetime* library and *strptime* function. Here is the code:

from datetime import datetime char_date = 'Apr 1 2015 1:20 PM' #creating example character date date_obj = datetime.strptime(char_date, '%b %d %Y %I:%M%p') print date_obj

I want to transpose Table A into Table B on the variable Product. This task can be accomplished by using Pandas **dataframe.pivot:**

Code

#Transposing Pandas dataframe by a variable

df=pd.read_excel("E:/transpose.xlsx", "Sheet1") # Load Data sheet of excel file EMP print df result= df.pivot(index= 'ID', columns='Product', values='Sales') result

**Output**

Data can be sorted using dataframe.sort() for data exploration in Python. It can be based on multiple variables and can be ascending or descending in both orders.

**Code**

#Sorting Pandas Dataframe df=pd.read_excel("E:/transpose.xlsx", "Sheet1") #Add by variable name(s) to sort print df.sort(['Product','Sales'], ascending=[True, False])

Above is a table with variables ID, Product, and Sales. Now, we want to sort it by Product and Sales (in descending order), as shown in Table 2.

Data visualization helps us understand the data easily. Python has libraries like matplotlib and Seaborn to create multiple graphs effectively. Let’s look at some of the visualizations below to understand the behavior of the variable(s).

- The distribution of age
- Relation between age and sales
- If sales are normally distributed or not?

**Code**

#Plot Histogram

import matplotlib.pyplot as plt import pandas as pd

df=pd.read_excel("E:/First.xlsx", "Sheet1")

#Plots in matplotlib reside within a figure object, use plt.figure to create new figure fig=plt.figure()

#Create one or more subplots using add_subplot, because you can't create blank figure ax = fig.add_subplot(1,1,1)

#Variable ax.hist(df['Age'],bins = 5)

#Labels and Tit plt.title('Age distribution') plt.xlabel('Age') plt.ylabel('#Employee') plt.show()

**Output**

**Code**

#Plots in matplotlib reside within a figure object, use plt.figure to create new figure fig=plt.figure()

#Create one or more subplots using add_subplot, because you can't create blank figure ax = fig.add_subplot(1,1,1)

#Variable ax.scatter(df['Age'],df['Sales'])

#Labels and Tit plt.title('Sales and Age distribution') plt.xlabel('Age') plt.ylabel('Sales') plt.show()

**Output**

**Code**

import seaborn as sns sns.boxplot(df['Age']) sns.despine()

**Output**

Frequency tables can be used to understand the distribution of categorical or n categorical variables using frequency tables.

**Code**

import pandas as pd

df=pd.read_excel("E:/First.xlsx", "Sheet1") print df

test= df.groupby(['Gender','BMI']) test.size()

**Output**

To select a sample from a data set, we will use the libraries numpy and random for data exploration in Python. Sampling data sets always helps us understand them quickly.

Let’s say, from the EMP table, I want to select a random sample of 5 employees.

**Code**

#Create Sample dataframe

import numpy as np import pandas as pd from random import sample

# create random index rindex = np.array(sample(xrange(len(df)), 5))

# get 5 random rows from the dataframe df dfr = df.ix[rindex] print dfr

**Output**

We often encounter duplicate observations. We can use the dataframe.drop_duplicates() to tackle this in Python.

**Code**

#Remove Duplicate Values based on values of variables "Gender" and "BMI"

rem_dup=df.drop_duplicates(['Gender', 'BMI']) print rem_dup

**Output**

I suggest you use a dataframe to understand the count, average, and sum of variables. Describe () with Pandas groupby() in data exploration in Python.

Let’s look at the code:

**Code**

test= df.groupby(['Gender']) test.describe()

**Output**

To identify missing values , we can use dataframe.isnull(). You can also refer article “Data Munging in Python (using Pandas)“, here we have done a case study to recognize and treat missing and outlier values.

**Code**

```
# Identify missing values of dataframe
df.isnull()
```

**Output**

Various imputation methods are available to treat missing values. You can refer these articles for methods to detect Outlier and Missing values. Imputation methods for both missing and outlier values are almost similar. Here, we will discuss general case imputation methods to replace missing values. Let’s do it using an example:

#Example to impute missing values in Age by the mean import numpy as np meanAge = np.mean(df.Age) #Using numpy mean function to calculate the mean value df.Age = df.Age.fillna(meanAge) #replacing missing values in the DataFrame

Joining/merging is a common operation to integrate datasets from different sources for data exploration in Python. They can be handled effectively in Pandas using the merge function:

**Code:**

df_new = pd.merge(df1, df2, how = 'inner', left_index = True, right_index = True) # merges df1 and df2 on index # By changing how = 'outer', you can do outer join. # Similarly how = 'left' will do a left join # You can also specify the columns to join instead of indexes, which are used by default.

This comprehensive guide examined the Python codes for various data exploration in Python and munging steps. We also examined the Python libraries Pandas, Numpy, Matplotlib, and Seaborn to perform these steps. In the next article, I will reveal the codes to perform these steps in R.

A. Data exploration in Python involves using libraries like Pandas for data manipulation, Matplotlib and Seaborn for visualization, and NumPy for numerical operations. It includes loading data, examining data types, summary statistics, missing values, correlations, and distributions to understand data structure and detect patterns or anomalies.

A. The data exploration process involves collecting and analyzing data to understand its structure, quality, and patterns. Steps include data collection, cleaning (handling missing values and outliers), summarizing with descriptive statistics, visualizing distributions, relationships, and trends, and deriving initial insights for further analysis.

A. Exploratory Data Analysis (EDA) in Python entails using tools like Pandas, Matplotlib, and Seaborn to summarize data sets visually and statistically. It involves examining data distributions, spotting outliers, detecting patterns, and checking assumptions to form hypotheses and guide data preprocessing and modeling.

A. The EDA process includes several steps: data collection, Data cleaning (handling missing values and outliers), Data transformation, Generating summary statistics, Visualizing data distributions and relationships, Drawing initial conclusions, and Identifying the next steps for modeling or further analysis.

Need the same thing/article in R. Please do need full asap. Thanks.

Hi Vikas, Guide for Data Exploration using R is live now. You can access it here http://www.analyticsvidhya.com/blog/2015/04/comprehensive-guide-data-exploration-r/ Thanks

About data imputation... I'm struggling to find methods (read packages / libraries) to carry out imputation in Python beyond the usual mean/ median approach...coming from an R background , there are tonnes of good packages to carry out very sophisticated imputation...this is something I miss in Python (it may be my lack of knowledge). i tried r2py to call imputation packages from R in python but it doesnt run as smoothly do you have suggestions in this regards?

Fantastic article with great examples! Thank you.