Installing NumPy Using pip
The NumPy module is not included in the standard Python distribution. However, it can be easily installed using the Python package installer pip:
Using Binary Packages
The best way to enable NumPy is to use prebuilt binary packages specific to your operating system. These packages include the full SciPy stack, which consists of NumPy, SciPy, Matplotlib, IPython, SymPy, and nose packages, in addition to the core Python.
Windows
- Anaconda: It is a free Python distribution that includes the entire SciPy stack. It is available for Windows, Linux, and macOS. Anaconda simplifies package management and comes with a convenient package installer called conda.
- Canopy: It is another distribution option available as both a free and commercial offering. Similar to Anaconda, Canopy provides a comprehensive SciPy stack for Windows, Linux, and macOS platforms.
- Python (x,y): is a free Python distribution for Windows OS. It includes the SciPy stack along with the Spyder IDE, making it suitable for scientific computing and data analysis tasks on Windows.
Linux
On Linux distributions, package managers are commonly used to install components of the SciPy stack. Here are examples for popular distributions:
To install Numpy on Ubuntu, following are the commands:
To install Numpy on Fedora, following are the commands:
Building NumPy from Source
For advanced users or those needing specific configurations, NumPy can be built from source. Ensure you have the necessary prerequisites:
- Core Python (2.6.x, 2.7.x, and 3.2.x onwards) with distutils enabled.
- zlib module enabled.
- GNU gcc (version 4.2 and above) C compiler.
To build and install NumPy from source, run the following command:
Verifying Installation
To verify that NumPy is installed correctly, open a Python prompt and try importing it:
If it is not installed, the following error message will be displayed:
Alternatively, you can import the NumPy package using the standard alias:
ImportError will indicate that NumPy is not found.
Checking NumPy Version
To check the version of NumPy installed on your system, you can use Python directly, either from the command line or within a script. Here is how you can do it:
Using Python Command Line or Script
- Open your command line interface (CLI) or terminal.
- Start Python by typing
pythonorpython3depending on your system configuration. - Import NumPy and check its version using the following commands:
print(np.__version__)
The above command will print the version of NumPy installed on your system.
Checking NumPy Version in a Script
Alternatively, you can create a Python script (e.g., check_numpy_version.py) with the following content:
print(f"NumPy version: {np.__version__}")
Save the file, then run it from the command line:
This script will output the NumPy version installed on your system.