Installing Python Data Science Packages on CentOS: Complete Step-by-Step Guide

Setting up a Python data science environment can be daunting for beginners, especially when dealing with package dependencies and system-specific installations. This comprehensive guide walks you through installing the three most essential Python data science packages on CentOS: NumPy, Pandas, and Scikit-learn.

🐍 Step 1: Checking Your Python Environment

Before installing any packages, it's crucial to verify your Python installation and version.

Checking Python Version

python --version

Output:

Python 3.9.23

What This Tells Us:

• Python 3.9.23 is installed and accessible
• This version is compatible with all modern data science packages
• The python command points to Python 3 (good for modern systems)

📦 Step 2: Installing pip (Python Package Manager)

Most CentOS systems don't come with pip pre-installed. Let's install it first.

Initial pip Installation Attempt

pip install numpy

Output:

bash: pip: command not found...
Install package 'python3-pip' to provide command 'pip'? [N/y] y

What Happened:

• The system detected that pip is not installed
• CentOS helpfully suggested installing python3-pip
• We accepted the installation prompt by typing y

System Package Installation Process

After accepting the pip installation, the system goes through several phases:

 * Waiting in queue...
 * Loading list of packages....
The following packages have to be installed:
 python3-pip-21.3.1-1.el9.noarch	A tool for installing and managing Python3 packages
Proceed with changes? [N/y] y

Installation Phases:

🔢 Step 3: Installing NumPy

With pip now available, the original NumPy installation command automatically continues:

Defaulting to user installation because normal site-packages is not writeable
Collecting numpy
  Downloading numpy-2.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (19.5 MB)
     |████████████████████████████████| 19.5 MB 3.4 MB/s
Installing collected packages: numpy
Successfully installed numpy-2.0.2

Understanding the NumPy Installation Output

Key Information:

1. User Installation Notice:

   Defaulting to user installation because normal site-packages is not writeable
   

   Copy
   • Packages install to user directory (~/.local/lib/python3.9/site-packages)
   • No admin privileges required for user-specific installations
   • Safer than system-wide installations

2. Package Download Details:

   • Version: numpy-2.0.2
   • Python Compatibility: cp39 (CPython 3.9)
   • Architecture: x86_64 (64-bit Linux)
   • File Size: 19.5 MB
   • Download Speed: 3.4 MB/s

3. File Format Explanation:

   • .whl = Wheel format (pre-compiled Python package)
   • manylinux = Compatible with many Linux distributions
   • Faster installation than compiling from source

📊 Step 4: Installing Pandas

Next, we install Pandas, which builds upon NumPy:

pip install pandas

Output:

Defaulting to user installation because normal site-packages is not writeable
Collecting pandas
  Downloading pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.4 MB)
     |████████████████████████████████| 12.4 MB 6.1 MB/s
Collecting tzdata>=2022.7
  Downloading tzdata-2025.2-py2.py3-none-any.whl (347 kB)
     |████████████████████████████████| 347 kB 38.7 MB/s
Requirement already satisfied: numpy>=1.22.4 in /home/centos9/.local/lib/python3.9/site-packages (from pandas) (2.0.2)
Collecting python-dateutil>=2.8.2
  Downloading python_dateutil-2.9.0.post0-py2.py3-none-any.whl (229 kB)
     |████████████████████████████████| 229 kB 21.1 MB/s
Collecting pytz>=2020.1
  Downloading pytz-2025.2-py2.py3-none-any.whl (509 kB)
     |████████████████████████████████| 509 kB 38.3 MB/s
Requirement already satisfied: six>=1.5 in /usr/lib/python3.9/site-packages (from python-dateutil>=2.8.2->pandas) (1.15.0)
Installing collected packages: tzdata, pytz, python-dateutil, pandas
Successfully installed pandas-2.3.2 python-dateutil-2.9.0.post0 pytz-2025.2 tzdata-2025.2

Understanding Pandas Installation and Dependencies

Main Package:

• pandas-2.3.2: The core data manipulation library (12.4 MB)
• Download Speed: 6.1 MB/s (faster than NumPy due to better network conditions)

Dependencies Installed:

Already Satisfied Dependencies:

• numpy>=1.22.4: Previously installed (2.0.2 satisfies requirement)
• six>=1.5: System package already available

🤖 Step 5: Installing Scikit-learn

Finally, we install Scikit-learn for machine learning capabilities:

pip install scikit-learn

Output:

Defaulting to user installation because normal site-packages is not writeable
Collecting scikit-learn
  Downloading scikit_learn-1.6.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.5 MB)
     |████████████████████████████████| 13.5 MB 19.4 MB/s
Collecting scipy>=1.6.0
  Downloading scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (38.6 MB)
     |████████████████████████████████| 38.6 MB 350 kB/s
Collecting threadpoolctl>=3.1.0
  Downloading threadpoolctl-3.6.0-py3-none-any.whl (18 kB)
Requirement already satisfied: numpy>=1.19.5 in /home/centos9/.local/lib/python3.9/site-packages (from scikit-learn) (2.0.2)
Collecting joblib>=1.2.0
  Downloading joblib-1.5.2-py3-none-any.whl (308 kB)
     |████████████████████████████████| 308 kB 13.3 MB/s
Installing collected packages: threadpoolctl, scipy, joblib, scikit-learn
Successfully installed joblib-1.5.2 scikit-learn-1.6.1 scipy-1.13.1 threadpoolctl-3.6.0

Understanding Scikit-learn Installation

Main Package:

• scikit-learn-1.6.1: Machine learning library (13.5 MB)
• Excellent Download Speed: 19.4 MB/s

Major Dependencies:

Notable Observations:

• SciPy is the largest package (38.6 MB) due to compiled mathematical algorithms
• Download speed variation: SciPy downloaded at only 350 kB/s (network congestion or server load)
• NumPy dependency satisfied: Our previously installed NumPy 2.0.2 meets the >=1.19.5 requirement

📋 Installation Summary

Let's summarize what we've accomplished:

Packages Successfully Installed

Total Resources Used

• Total Download Size: ~85 MB across all packages
• Installation Location: ~/.local/lib/python3.9/site-packages/
• Installation Type: User-level (no admin privileges required)

🎯 Key Takeaways

🚀 What's Next?

Now that you have the essential data science packages installed, you're ready to start coding! In our next post, we'll explore:

• Creating and testing simple scripts with these packages
• Understanding basic NumPy array operations
• Working with Pandas DataFrames
• Loading datasets with Scikit-learn
• Practical examples and code demonstrations

The foundation is set – let's start building amazing data science projects!

🔧 Troubleshooting Tips

Common Issues and Solutions:

• Permission Denied: Use --user flag with pip for user installations
• Command Not Found: Ensure pip is installed and accessible in your PATH
• Version Conflicts: Use virtual environments to isolate package versions
• Slow Downloads: Try different times of day or use pip with --index-url flag

💬 Discussion

Have you encountered any issues during your Python data science setup?

• Which package took the longest to install on your system?
• Have you tried installing these packages on other Linux distributions?
• What data science projects are you planning to work on?
• Did you prefer user installation or would you use virtual environments?

Connect with me:

• 🐙 GitHub - Data science examples and scripts
• 🐦 Twitter - Quick tips and updates
• 📧 Contact - Data science discussions and questions

This installation guide covers the fundamental setup process for Python data science packages. As package versions evolve, some details may change, but the core installation principles remain constant.