SciPy and NumPy

SciPy is a collection of open source code libraries for math, science and engineering. NumPy, Matplotlib and pandas are libraries that fall under the SciPy project umbrella.

NumPy (source code) is a Python code library that adds scientific computing capabilities such as N-dimensional array objects, FORTRAN and C++ code integration, linear algebra and Fourier transformations. NumPy serves as a required dependency for many other scientific computing packages such as pandas.

Blaze is a similar, but separate, ecosystem with additional tools for wrangling, cleaning, processing and analyzing data.

SciPy resources

Take a look at the individual pages for NumPy, Matplotlib and pandas for tutorials specific to those projects. The following resources are broader walkthroughs for the SciPy ecosystem:

• SciPy Lecture notes goes into the overall Python scientific computing ecosystem and how to use it.

• The SciPy Cookbook contains instructions for various SciPy packages that were previously hosted on the SciPy wiki.

• Lectures in Quantitative Economics: SciPy provides a good overview of SciPy compared to the specific NumPy project, as well as explanations for the wrappers SciPy provides over lower-level FORTRAN libraries.

• A plea for stability in the SciPy ecosystem presents concerns from one scientist's perspective about how fast the Python programming ecosystem changes and that code can become backwards incompatible in only a few years. The issue is that many science projects last decades and therefore cannot follow the rate of change as easily as typical software development projects.

NumPy resources

• From Python to NumPy is an awesome resource that shows how to use your basic Python knowledge to learn how to do vectorization with NumPy.

• 101 NumPy Exercises for Data Analysis

• NumPy: creating and manipulating numerical data

• Python NumPy Array Tutorial is a starter tutorial specifically focused on using and working with NumPy's powerful arrays.

• Beyond Numpy Arrays in Python is a predecessor to a Numpy Enhancement Proposal that recommends how to prepare the scientific computing ecosystme for GPU, distributed and sparse arrays.

Example NumPy code

• SmoothLife is an implementation of Conway's Game of Life using NumPy. The project uses a continuous space rather than the traditional discrete board.

您还想了解有关 Python 和数据的其他信息？