.. _chap_installation:

Installation
============


In order to get you up and running for hands-on learning experience, we
need to set you up with an environment for running Python, Jupyter
notebooks, the relevant libraries, and the code needed to run the book
itself.

Installing Miniconda
--------------------

The simplest way to get going will be to install
`Miniconda <https://conda.io/en/latest/miniconda.html>`__. The Python
3.x version is required. You can skip the following steps if your
machine already has conda installed.

Visit the Miniconda website and determine the appropriate version for
your system based on your Python 3.x version and machine architecture.
For example, if you are using macOS and Python 3.x you would download
the bash script whose name contains the strings "Miniconda3" and
"MacOSX", navigate to the download location, and execute the
installation as follows (taking Intel Macs as an example):

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    sh Miniconda3-py39_4.12.0-MacOSX-x86_64.sh -b

A Linux user with Python 3.x would download the file whose name contains
the strings "Miniconda3" and "Linux" and execute the following at the
download location:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    sh Miniconda3-latest-Linux-x86_64.sh -b

Next, initialize the shell so we can run ``conda`` directly.

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    ~/miniconda3/bin/conda init

Now close and reopen your current shell. You should be able to create a
new environment as follows:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    conda create --name d2l python=3.9 -y

Downloading the D2L Notebooks
-----------------------------

Next, we need to download the code of this book. You can click the "All
Notebooks" tab on the top of any HTML page to download and unzip the
code. Alternatively, if you have ``unzip`` (otherwise run
``sudo apt install unzip``) available:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    mkdir d2l-en && cd d2l-en
    curl https://d2l.ai/d2l-en.zip -o d2l-en.zip
    unzip d2l-en.zip && rm d2l-en.zip

Now we can activate the ``d2l`` environment:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    conda activate d2l

Installing the Framework and the ``d2l`` Package
------------------------------------------------

Before installing any deep learning framework, please first check
whether or not you have proper GPUs on your machine (the GPUs that power
the display on a standard laptop are not relevant for our purposes). If
you are working on a GPU server, proceed to :ref:`subsec_gpu` for
instructions on how to install GPU-friendly versions of the relevant
libraries.

If your machine does not house any GPUs, there is no need to worry just
yet. Your CPU provides more than enough horsepower to get you through
the first few chapters. Just remember that you will want to access GPUs
before running larger models. To install the the CPU version, execute
the following command.



.. raw:: html

     <div class="mdl-tabs mdl-js-tabs mdl-js-ripple-effect"><div class="mdl-tabs__tab-bar text"><a href="#mxnet-1-0" onclick="tagClick('mxnet'); return false;" class="mdl-tabs__tab is-active">mxnet</a><a href="#pytorch-1-1" onclick="tagClick('pytorch'); return false;" class="mdl-tabs__tab ">pytorch</a><a href="#tensorflow-1-2" onclick="tagClick('tensorflow'); return false;" class="mdl-tabs__tab ">tensorflow</a></div>



.. raw:: html

     <div class="mdl-tabs__panel is-active" id="mxnet-1-0">

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    pip install mxnet==1.7.0.post1



.. raw:: html

     </div>



.. raw:: html

     <div class="mdl-tabs__panel " id="pytorch-1-1">

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    pip install torch torchvision



.. raw:: html

     </div>



.. raw:: html

     <div class="mdl-tabs__panel " id="tensorflow-1-2">

You can install TensorFlow with both CPU and GPU support as follows:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    pip install tensorflow tensorflow-probability



.. raw:: html

     </div>



.. raw:: html

     </div>

Our next step is to install the ``d2l`` package that we developed in
order to encapsulate frequently used functions and classes found
throughout this book.

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    pip install d2l==0.17.5

Once you have completed these installation steps, we can the Jupyter
notebook server by running:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    jupyter notebook

At this point, you can open http://localhost:8888 (it may have already
opened automatically) in your Web browser. Then we can run the code for
each section of the book. Please always execute ``conda activate d2l``
to activate the runtime environment before running the code of the book
or updating the deep learning framework or the ``d2l`` package. To exit
the environment, run ``conda deactivate``.

.. _subsec_gpu:

GPU Support
-----------




.. raw:: html

     <div class="mdl-tabs mdl-js-tabs mdl-js-ripple-effect"><div class="mdl-tabs__tab-bar text"><a href="#mxnet-3-0" onclick="tagClick('mxnet'); return false;" class="mdl-tabs__tab is-active">mxnet</a><a href="#pytorch-3-1" onclick="tagClick('pytorch'); return false;" class="mdl-tabs__tab ">pytorch</a><a href="#tensorflow-3-2" onclick="tagClick('tensorflow'); return false;" class="mdl-tabs__tab ">tensorflow</a></div>



.. raw:: html

     <div class="mdl-tabs__panel is-active" id="mxnet-3-0">

By default, MXNet is installed without GPU support to ensure that it
will run on any computer (including most laptops). Part of this book
requires or recommends running with GPU. If your computer has NVIDIA
graphics cards and has installed
`CUDA <https://developer.nvidia.com/cuda-downloads>`__, then you should
install a GPU-enabled version. If you have installed the CPU-only
version, you may need to remove it first by running:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    pip uninstall mxnet

We now need to find out what version of CUDA you have installed. You can
check this by running ``nvcc --version`` or
``cat /usr/local/cuda/version.txt``. Assume that you have installed CUDA
10.2, then you can install with the following command:

.. raw:: latex

   \diilbookstyleinputcell

.. code:: bash

    # For Windows users
    pip install mxnet-cu102==1.7.0 -f https://dist.mxnet.io/python

    # For Linux and macOS users
    pip install mxnet-cu102==1.7.0

You may change the last digits according to your CUDA version, e.g.,
``cu101`` for CUDA 10.1 and ``cu90`` for CUDA 9.0.



.. raw:: html

     </div>



.. raw:: html

     <div class="mdl-tabs__panel " id="pytorch-3-1">

By default, the deep learning framework is installed with GPU support.
If your computer has NVIDIA GPUs and has installed
`CUDA <https://developer.nvidia.com/cuda-downloads>`__, then you are all
set.



.. raw:: html

     </div>



.. raw:: html

     <div class="mdl-tabs__panel " id="tensorflow-3-2">

By default, the deep learning framework is installed with GPU support.
If your computer has NVIDIA GPUs and has installed
`CUDA <https://developer.nvidia.com/cuda-downloads>`__, then you are all
set.



.. raw:: html

     </div>



.. raw:: html

     </div>

Exercises
---------

1. Download the code for the book and install the runtime environment.



.. raw:: html

     <div class="mdl-tabs mdl-js-tabs mdl-js-ripple-effect"><div class="mdl-tabs__tab-bar text"><a href="#mxnet-5-0" onclick="tagClick('mxnet'); return false;" class="mdl-tabs__tab is-active">mxnet</a><a href="#pytorch-5-1" onclick="tagClick('pytorch'); return false;" class="mdl-tabs__tab ">pytorch</a><a href="#tensorflow-5-2" onclick="tagClick('tensorflow'); return false;" class="mdl-tabs__tab ">tensorflow</a></div>



.. raw:: html

     <div class="mdl-tabs__panel is-active" id="mxnet-5-0">

`Discussions <https://discuss.d2l.ai/t/23>`__



.. raw:: html

     </div>



.. raw:: html

     <div class="mdl-tabs__panel " id="pytorch-5-1">

`Discussions <https://discuss.d2l.ai/t/24>`__



.. raw:: html

     </div>



.. raw:: html

     <div class="mdl-tabs__panel " id="tensorflow-5-2">

`Discussions <https://discuss.d2l.ai/t/436>`__



.. raw:: html

     </div>



.. raw:: html

     </div>