Installing Intel® Distribution for Python* and Intel® Performance Libraries with Anaconda*

We have worked with Continuum Analytics* to make it easy to use Intel® Distribution for Python and the Intel® Performance Libraries (such as Intel® Math Kernel Library (Intel® MKL)) with the Conda* package manager and Anaconda Cloud*. You need at least conda 4.1.11, so first update your conda.

conda update conda

Tell conda to choose Intel packages over default packages, when available.

conda config --add channels intel

Installing the Intel® Distribution for Python*

We recommend that you create a new environment when installing. To install the core python3 environment, do:

conda create -n idp intelpython3_core python=3

If you want python 2 do:

conda create -n idp intelpython2_core python=2

If you want the full Intel distribution, replace the "core" package name with "full", like this for python3:

conda create -n idp intelpython3_full python=3

Then follow the usual directions for activating the environment. Linux/macOS users do:

source activate idp

and Microsoft Windows users do:

activate idp

You now have the core environment, including python, numpy, scipy,... You can use the usual conda install commands for additional packages. For example, to install intel sympy do:

conda install sympy

Non-intel packages are installed as usual. For example, to install affine do:

conda install affine

Available Intel packages can be viewed here: https://anaconda.org/intel/packages

Using Intel Conda* Packages with Continuum's Python*

If you want to install Intel packages into an environment with Continuum's python, do not add the "intel" channel to your configuration file because that will cause all your Continuum packages to be replaced with Intel builds, if available. Rather, specify the "intel" channel on the command line with "-c intel" parameter and the "--no-update-deps" flag to avoid switching other packages, such as python itself, to Intel's builds:

conda install mkl -c intel --no-update-deps
conda install numpy -c intel --no-update-deps

Installing the Intel® Performance Libraries

If you want to build a native extension that directly uses the performance libraries, then you will need to obtain a development package that contains header files and static libraries. We have published them as conda packages for your convenience. 

Make sure the Intel channel is added to your conda configuration (see above). Then install any of our available performance libraries using "conda install" as normal, such as:

conda install mkl-devel

The following table lists the available packages with a brief description for their contents:

Package NameLin‑64Lin‑32Win‑64Win‑32macOS‑64Description
mklXXXXXIntel® Math Kernel Library (Intel® MKL) dynamic runtimes
mkl‑develXXXXXIntel® MKL dynamic runtimes and headers for building software
mkl‑staticXXXXXIntel® MKL static libraries and headers for building software
mkl‑includeXXXXXIntel® MKL headers only. Automatically installed along with development packages
       
       

 

For more complete information about compiler optimizations, see our Optimization Notice.

9 comments

Top
Vikrant K. (Intel)'s picture

Hi

I am getting the following error when creating the environment and installing the package.

Solving environment: failedCondaHTTPError: HTTP 000 CONNECTION FAILED for url <https://repo.anaconda.com/pkgs/pro/noarch/repodata.json.bz2>
Elapsed: -An HTTP error occurred when trying to retrieve this URL.
HTTP errors are often intermittent, and a simple retry will get you on your way.If your current network has https://www.anaconda.com blocked, please file
a support request with your network engineering team. ConnectTimeout(MaxRetryError("HTTPSConnectionPool(host='repo.anaconda.com', port=443): Max retries exceeded with url: /pkgs/pro/noarch/repodata.json.bz2 (Caused by ConnectTimeoutError(<urllib3.connection.VerifiedHTTPSConnection object at 0x00000299626090B8>, 'Connection to repo.anaconda.com timed out. (connect timeout=9.15)'))"))

 

I have seen this error mentioned on GitHub and stackoverflow as well, but it's not clear what the solution is.

Can anyone suggest how to resolve it?

Thanks

M. Mizuno's picture

What happened to mkl and mkl-devel? The pages are private now.
https://anaconda.org/intel/mkl
https://anaconda.org/intel/mkl-devel

I also found that neither mkl=2019.0=118 -c intel nor mkl-devel=2019.0=118 -c intel installs import libraries (such as mkl_core_dll.lib) on Windows (64bit).

G Anthony R. (Intel)'s picture

Does the current Anaconda distribution use this by default now? Or do I still need to specify the intelpython3 from the Intel channel?

Tabish's picture

@kumar,lalit try again with this 

conda create -n idp intelpython3_core python=3

and then do source activate idp. I think all the libraries and dependencies are not installed therefore it is causing a problem. 

Robert C. (Intel)'s picture

Here is more info on installing using pip: https://software.intel.com/en-us/articles/installing-the-intel-distribution-for-python-and-intel-performance-libraries-with-pip-and

 

Royi's picture

Could you please distribute them through PIP as well?

Thank You.

kumar, lalit's picture

I am getting this Error After hitting command in Anaconda prompt(Windows) "activate idp":  'chcp' is not recognized as an internal or external command,operable program or batch file.

Domingos, Diogenes A.'s picture

the advise If you want to install Intel packages into an environment with Continuum's python, do not add the "intel" should stay before the command to install of intel channels

Add a Comment

Have a technical question? Visit our forums. Have site or software product issues? Contact support.