Scientific computing packages such as Python's scipy installed under Windows

If you want to use Python for scientific calculations, data analysis, etc., be sure to install scipy, Seaborn, NumPy, and so on.

However, there are often problems with third-party libraries that install Python under Windows. Previously, it has been described how to install lxml under Windows, in fact, this installation method is also suitable for other scenarios that cannot be installed through PIP or the IDE Package management tool.

If the main purpose of using Python is for scientific computing, data analysis, crawlers and other projects, it is highly recommended to use Anaconda instead of other Ides.

In addition to the Python language, Anaconda integrates scientific computing packages such as NumPy, SciPy, Matplotlib, and other network-related packages such as Beautiful-soup, requests, and lxml.

After installing Anaconda, there is no need to bother installing additional third-party libraries.

A common installation method that can be applied to other Ides and other third-party libraries is as follows:

First, make sure that the wheel library is also installed outside of the PIP. If not: Open cmd, enter pip install wheel to install

This method installs a third-party library from a binary file and requires that you determine the version that can be installed before you install it.

The confirmation method is as follows:

In cmd, enter python into Python (also available in the IDE).

Then enter import Pip;print (pip.pep425tags.get_supported ()) to output the current Python version information on the interface.

Find the appropriate WHL file on the www.lfd.uci.edu/%7Egohlke/pythonlibs/, download it locally. Note In the version information that is marked, the corresponding version should be downloaded from the Web page: XXXX-CP35-CP35M-WIN_AMD64.WHL

Here, the scipy corresponding installation package is SCIPY?0.18.1?CP35?CP35M?WIN_AMD64.WHL (please replace with your installable version). Note that scipy relies on NUMPY+MKL to install the NUMPY+MKL first before installing SCIPY.

Even if you have previously installed NumPy, please find the NUMPY+MKL WHL from this page and download it locally, because the focus is on MKL rather than numpy.

Here's how to install the Library through WHL:

Go to the folder where the WHL file is downloaded, press and hold down the SHIFT key while clicking the right mouse button and select "Open Command Window Here" in the pop-up menu.

In the command window that opens, enter the full name of the PIP install +WHL file, and install the WHL.

Install NUMPY+MKL, enter pip install NUMPY?1.12.0B1+MKL?CP35?CP35M?WIN_AMD64.WHL, press ENTER, and wait for the installation to complete. (note replace the filename with the name of your available version)

Install SCIPY, enter pip install SCIPY?0.18.1?CP35?CP35M?WIN_AMD64.WHL, press ENTER, and wait for the installation to complete. (note replace the filename with the name of your available version)

Https://jingyan.baidu.com/album/ca41422f27c56a1eae99ed39.html?picindex=3

Scientific computing packages such as Python's scipy installed under Windows (GO)