The first lesson several environment configurations of Python

First, Pythom+eclipse+pydev.

This kind of installation is relatively simple, the online tutorial is more, it is important to note that installation of the JDK before eclipse needs to install. The concrete process is no longer verbose. The following is mainly about the 64-bit system installed Numpy,scipy,matplotlib, such as several scientific calculation package.

Python's use of the two libraries, NumPy and scipy, is also useful in scientific calculations, but the problem is that the two libraries are not well-fitted, especially in 64-bit situations.

The official PyPI on the default only 32-bit, and SourceForge on the Windows EXE installation package (seemingly) there is no 64-bit.

I am under 2.7 Direct pip install cannot install numpy,3.4 under numpy Normal installation, but scipy is dead or dead installed.

Fortunately, at the University of California, Irvine (University of California, Irvine) LFD Laboratory (the Laboratory for fluorescence Dynamics) Christoph Gohlke has an unofficial Windows source on its profile: http://www.lfd.uci.edu/~gohlke/pythonlibs/.

Just go to download the corresponding version of the installation is good, you need to pay attention to the dependencies.

Delbert himself installs the 64-bit version of Python 2.7 and 3.4, which defaults to the 3.4 version. As more and more Python libraries are encapsulated in wheel, that is, the extension becomes. WHL, and cannot be directly double-clicked as the previous. exe. Well done.

Python 2.7 AMD64 installation numpy and scipy

Because the downloaded package is in wheel format, wheel support is required. The default version of Delbert is not 2.7, so you need to switch to the script folder under the 2.7 installation path by using the CD command, and enter PIP install wheel to install wheel.

Then pip install < download to the NumPy installation package > to install it.

After installing NumPy, it is also installed via Pip install < download to the SCIPY installation package >.

If you are in a Unix-like environment, the path needs to be the absolute path to/from the beginning, otherwise you will be prompted not to find the file error.

Python 3.4 AMD64 installation NumPy and scipy

Since the default version of Delbert is 3.4, the installation of version 3.4 is relatively easy.

Open a command line, pip install Wheel install wheel.

Pip Install < Drag into the downloaded NumPy package >

Pip Install < Drag into the downloaded scipy package >

Well, then the installation is done.

At the command line, you do not need to manually enter the Unix-like path, just drag the installation package into the command-line window to get the full path.

Precautions

It seems that numpy and official conflicts from unofficial sources need to be manually uninstalled if they were manually installed, otherwise the uninstallation will fail to complete the installation while the NumPy is being installed.

Test it, no problem, start working.

Above reprinted from Delbert's Blog .

Special attention should be paid to the location of NumPy and so on, if it is not well put in/python/scripts inside it.

The second kind,Anaconda

It is said that the first method is a big pit, but the process of installation to understand Python still has a certain role, no way, I like to toss ...

This software I also installed, one-click installation, easy to solve, but still do not understand good this software, just feel very cow look, online a lot of said note will not go, anyway I was on the go, haha.

The Third Kind, Winpython

This did not go to, do not want to go to, today enough toss.

Above is a summary of today's toss, I will choose the first way to learn Python, slowly start to choose a more comfortable software bar. But the feeling is Linux system Cow b AH.

Please disregard my logic and expression. Ha ha haha

Several environment configurations for Python in the first lesson