:orphan:

|rss_image|  **install - 1/2** :ref:`==> <ap-cat-install-1>` :ref:`Blog <ap-main-0>` :ref:`installation (12) <ap-cat-installation-0>` :ref:`python (3) <ap-cat-python-0>`

.. |rss_image| image:: feed-icon-16x16.png
    :target: ../_downloads/rss.xml
    :alt: RSS

----


.. index:: install


.. _ap-cat-install-0:

install - 1/2
+++++++++++++

.. blogpostagg::
    :title: Build xgboost 0.4a30 for Python 3.4 and 3.5
    :date: 2016-04-04
    :keywords: install,xgboost,vcomp110.dll,OpenMP
    :categories: install,module,windows
    :rawfile: 2016/2016-04-04_build_xgboost_04a30.rst

    `xgboost <https://github.com/dmlc/xgboost>`_ does not support the Windows
    compilation anymore so I had to look for a way to compile *xgboost* for Windows.
    I tried to start from the source but it seemed quite complex plus the owners
    mention they don't support it anymore. Instead, I chose to use
    the source included the release package on
    `PyPi <https://pypi.python.org/pypi/xgboost>`_.
    I copy paste some of the instructions I used from the previous
    blog post I wrote on that topic since I did not start from a fresh and empty
    environment. The steps I did not do again are marked with a ``*``.
    This is the process I followed:
    

    ...




.. blogpostagg::
    :title: Directly import from Github
    :date: 2016-01-05
    :keywords: github,import
    :categories: install,module
    :rawfile: 2016/2016-01-05_import_from_github.rst

    If you want to do something like::
    

    ...




.. blogpostagg::
    :title: pythonnet for Python 3.5
    :date: 2016-01-03
    :keywords: pythonnet,.net,C#,python,issue
    :categories: install,module,pythonnet
    :rawfile: 2016/2016-01-03_pythonnet.rst

    People usually install modules from
    `Unofficial Windows Binaries for Python Extension Packages <http://www.lfd.uci.edu/~gohlke/pythonlibs/#pythonnet>`_
    on Windows. This also goes for
    `pythonnet <https://github.com/fkarb/pythonnet>`_ which was compiled
    this github repository. This release works somehow for Python 3.5 but fails if
    some functionalities are required. You can try::
    

    ...




.. blogpostagg::
    :title: PyCrypto and pycryptodome
    :date: 2015-12-21
    :keywords: pycrypto,pycryptodome
    :categories: install,module
    :rawfile: 2015/2015-12-21_pycrypto.rst

    `pycrypto <https://pypi.python.org/pypi/pycrypto/>`_ is not longer maintained,
    it should be replaced by `pycryptodome <https://pypi.python.org/pypi/pycryptodome/>`_.
    The second module is a fork of the first.
    You can check on gihub
    `Legrandin/pycryptodome <https://github.com/Legrandin/pycryptodome>`_
    for the latest release.




.. blogpostagg::
    :title: Install a package on Anaconda
    :date: 2015-12-13
    :keywords: install
    :categories: install,module,Anaconda
    :rawfile: 2015/2015-12-13_conda.rst

    Some packages are specifically compiled for
    `Anaconda <https://www.continuum.io/downloads>`_ distribution.
    That's why it is better to install them with the command::
    

    ...




.. blogpostagg::
    :title: Wordcloud to draw clouds of words
    :date: 2015-12-05
    :keywords: install,wordcloud
    :categories: install,module
    :rawfile: 2015/2015-12-05_wordcloud.rst

    `wordcloud <https://github.com/amueller/word_cloud>`_
    is a nice module to draw `tag clouds <https://en.wikipedia.org/wiki/Tag_cloud>`_
    in Python. We can read the blog post
    `A Wordcloud in Python <http://peekaboo-vision.blogspot.de/2012/11/a-wordcloud-in-python.html>`_
    or see the `gallery <http://amueller.github.io/word_cloud/>`_.
    An example of what it can do:
    

    ...




.. blogpostagg::
    :title: A few interested modules not included in this one
    :date: 2015-12-05
    :keywords: install,wordcloud,ORM,images,cluster,HMM
    :categories: install,module
    :rawfile: 2015/2015-12-05_interesting_modules.rst

    The following might be included in a near future.
    

    ...




.. blogpostagg::
    :title: Pycrypto on Python 3.5
    :date: 2015-11-25
    :keywords: install,pycrypto
    :categories: install,module,windows
    :rawfile: 2015/2015-11-25_pycrypto.rst

    Fortunately, somebody already did that
    and it is available here:
    `sfbahr/PyCrypto-Wheels <https://github.com/sfbahr/PyCrypto-Wheels>`_.




.. blogpostagg::
    :title: Install Anaconda through SSH connection
    :date: 2015-11-01
    :keywords: Anaconda,SSH,Miniconda
    :categories: install,Anaconda,SSH
    :rawfile: 2015/2015-11-01_anaconda_ssh.rst

    The following page
    `Install/Uninstall Anaconda Server <http://docs.continuum.io/anaconda-server/easy-gui/index>`_
    describes how to install Anaconda through a SSH connection and only a command line.
    It starts from `Miniconda <http://conda.pydata.org/miniconda.html>`_
    which is Anaconda without this exhaustive list of modules.
    By default, the page gives the instruction for Python 2.7.
    I just changed them to get Python 3.4.
    

    ...




.. blogpostagg::
    :title: Install cvxpy on Windows
    :date: 2015-09-20
    :keywords: cvxpy,Windows
    :categories: install,module,issue
    :rawfile: 2015/2015-09-20_cvxpy.rst

    After I installed `CVXcanon <https://github.com/cvxgrp/CVXcanon>`_ (see previous blog),
    I went through an issue
    ``canonInterface`` is not found.
    I corrected with the following modification:
    `change import instruction to use CVXcanon explicitely  <https://github.com/sdpython/cvxpy/commit/11fe2bff21cbafcb694cab1265f111e6f9314ac4>`_.





----

|rss_image|  **install - 1/2** :ref:`==> <ap-cat-install-1>` :ref:`2019-11 (2) <ap-month-2019-11-0>` :ref:`2019-12 (1) <ap-month-2019-12-0>` :ref:`2020-02 (1) <ap-month-2020-02-0>` :ref:`2020-09 (1) <ap-month-2020-09-0>` :ref:`2021-01 (5) <ap-month-2021-01-0>`