All PostsColumns
ho-1
ho-1

5 Essential Python Libraries for Data Analysis

What should I prepare to start data analysis?

First, it is important to learn the basics of Python, which is the most widely used programming language for data analysis. You should also know the essential libraries that are often used. Among them, I will introduce the five most important libraries and review the characteristics, advantages, disadvantages, and simple codes of each library. Thanks to ChatGPT, you can use them sufficiently with just the basics, so shall we learn together?

📌 What is a library?

A library is like a superhero's utility belt. It pulls out the necessary functions to solve the mission smoothly. It's like starting with knowing what ingredients give what flavors when cooking.

1. Pandas

Pandas is an essential library for manipulating and analyzing data in Python. It can handle various file formats (CSV, Excel, SQL, etc.), making it a starting point for dealing with a wide range of data in Python.

Pandas: https://pandas.pydata.org/

Advantages

  • Designed to efficiently handle tabular data

  • Easy conversion to DataFrames

  • Intuitively performs operations such as filtering and aggregation

  • Efficient for processing large datasets

  • Excellent compatibility with various file formats (CSV, Excel, SQL, etc.) for easy data import and export

Disadvantages

  • Performance degrades when handling very large datasets as data is loaded into memory

  • Slower data processing speed compared to other C++ based tools

Example Code

# Importing the library
import pandas as pd 

# Creating a DataFrame
data = pd.DataFrame({
    'Name': ['Kim Cheolsu', 'Lee Younghee', 'Park Minsu'],
    'Age': [25, 30, 35],
    'Job': ['Student', 'Company Employee', 'Teacher']
})

# Reading data
data = pd.read_csv('data.csv')  # Reading CSV file
data = pd.read_excel('data.xlsx')  # Reading Excel file

# Basic operations
print(data.head())  # Viewing the top 5 rows
print(data.info())  # Viewing data information
print(data.describe())  # Viewing descriptive statistics

2. NumPy

NumPy is a Python library for numerical computations. It supports array and matrix operations and is a basic computational tool for tasks in data analysis, scientific computing, and machine learning.

NumPy: https://numpy.org/

Advantages

  • Provides multi-dimensional array objects

  • Useful for fast computations and handling large datasets

  • Supports vectorized operations, providing higher performance than loops

  • Includes essential functions for scientific computing such as linear algebra and Fourier transforms

Disadvantages

  • More difficult to use for complex data analysis compared to Pandas

  • Cannot handle labeled data like DataFrames

Example Code

# Importing the library
import numpy as np

# Creating an array
arr = np.array([1, 2, 3, 4, 5])
matrix = np.array([[1, 2, 3], [4, 5, 6]])

# Creating special arrays
zeros = np.zeros((3, 3))  # 3x3 matrix filled with 0
ones = np.ones((2, 2))    # 2x2 matrix filled with 1
random = np.random.rand(3, 3)  # Random value matrix

# Basic operations
print(arr + 2)  # Adding 2 to all elements
print(arr * 2)  # Multiplying all elements by 2

3. Matplotlib

Matplotlib is a Python library for data visualization. It can create various graphs such as line graphs, bar graphs, and scatter plots, and allows detailed adjustments to the style and elements of the graphs.

Matplotlib: https://matplotlib.org/

Advantages

  • Easy to implement simple visualizations; structured in a way that beginners can easily learn

Disadvantages

  • The basic styles of visualizations are somewhat monotonous

  • Code can become lengthy when drawing complex graphs

Example Code

# Importing the library
import matplotlib.pyplot as plt

# Basic graph
x = [1, 2, 3]
y = [2, 4, 6]
plt.plot(x, y)
plt.show()

Example of a chart image created with Matplotlib: https://matplotlib.org/stable/gallery/index.html

4. Seaborn

Seaborn is a data visualization library designed based on Matplotlib. It has a similar usage to Matplotlib but features more concise and intuitive code.

Seaborn: https://seaborn.pydata.org/

Advantages

  • Designed based on Matplotlib, so the usage is similar

  • Easily creates statistical graphs (distributions, correlation heatmaps, etc.)

  • Beautiful and intuitive default styles

  • Directly linked to DataFrames, making it easy to derive insights from complex datasets

Disadvantages

  • Customization is more limited compared to Matplotlib

Example Code

# Importing the library
import seaborn as sns

# Setting basic style
sns.set_style("whitegrid")

# Drawing a distribution plot
sns.distplot(df['column_name'])

# Heatmap
sns.heatmap(df.corr(), annot=True)

# Scatterplot matrix
sns.pairplot(df)

# Boxplot
sns.boxplot(x='category_column', y='numeric_column', data=df)

Example of a chart image created with Seaborn: https://seaborn.pydata.org/examples/index.html

5. Scikit-learn

Scikit-learn is a Python library for machine learning. It provides various algorithms and tools for simple and efficient data analysis and modeling, and is widely used from beginners to experts.

Scikit-learn: https://scikit-learn.org/stable/

Advantages

  • Supports the entire process of machine learning, including data preprocessing, model training, and evaluation

  • Includes various algorithms, making it highly practical for real-world use

  • Easy to apply multiple models such as classification, regression, and clustering

  • Rich tools for evaluating and improving model performance

Disadvantages

  • Requires separate libraries for handling complex models like deep learning

Example Code

# Importing the library
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error

# Splitting the data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)

# Training the model
model = LinearRegression()
model.fit(X_train, y_train)

# Making predictions
predictions = model.predict(X_test)

# Evaluating performance
mse = mean_squared_error(y_test, predictions)
print(f'MSE: {mse}')

We have looked at all the essential libraries needed for data analysis. I hope the organized information is well utilized and helpful.

📌 Summary

  1. Pandas: The basic tool for data manipulation and analysis, supports various file formats

  2. NumPy: Essential library for numerical computations and large-scale data processing

  3. Matplotlib: Creates various graphs and charts for data visualization

  4. Seaborn: Easily understands data distribution and relationships with advanced visualization

  5. Scikit-learn: A powerful tool for building and evaluating machine learning models

Python
Comment
No comment