Skip to content

Files

Latest commit

2e28fb6 · Dec 29, 2021

History

History
253 lines (165 loc) · 6.78 KB

INSTALL.rst.txt

File metadata and controls

253 lines (165 loc) · 6.78 KB

Building and installing SciPy

See https://www.scipy.org/install/

It is strongly recommended that you use either a complete scientific Python distribution or binary packages on your platform if they are available, in particular on Windows and Mac OS X. You should not attempt to build SciPy if you are not familiar with compiling software from sources.

Recommended distributions are:

The rest of this install documentation summarizes how to build Scipy. Note that more extensive (and possibly more up-to-date) build instructions are maintained at https://scipy.github.io/devdocs/building/

SciPy requires the following software installed for your platform:

  1. Python >= 3.8
  1. NumPy >= 1.18.5

If building from source, SciPy also requires:

  1. setuptools < 60.0
  1. pybind11 >= 2.4.3
  1. If you want to build the documentation: Sphinx >= 2.4.0 and < 3.1.0
  1. Cython >= 0.29.18

There are two ways to build SciPy on Windows:

  1. Use Intel MKL, and Intel compilers or ifort + MSVC. This is what Anaconda and Enthought Canopy use.
  2. Use MSVC + GFortran with OpenBLAS. This is how the SciPy Windows wheels are built.

It is recommended to use GCC or Clang, both work fine. Gcc is available for free when installing Xcode, the developer toolsuite on Mac OS X. You also need a Fortran compiler, which is not included with Xcode: you should use a recent GFortran from an OS X package manager (like Homebrew).

Please do NOT use GFortran from hpc.sourceforge.net, it is known to generate buggy SciPy binaries.

You should also use a BLAS/LAPACK library from an OS X package manager. ATLAS, OpenBLAS, and MKL all work.

As of SciPy version 1.2.0, we do not support compiling against the system Accelerate library for BLAS and LAPACK. It does not support a sufficiently recent LAPACK interface.

Most common distributions include all the dependencies. You will need to install a BLAS/LAPACK (all of ATLAS, OpenBLAS, MKL work fine) including development headers, as well as development headers for Python itself. Those are typically packaged as python-dev.

For the latest information, see the website:

https://www.scipy.org

Use the command:

git clone https://github.com/scipy/scipy.git

cd scipy
git clean -xdf
python setup.py install --user

Type:

cd scipy/doc
make html

Unpack SciPy-<version>.tar.gz, change to the SciPy-<version>/ directory, and run:

pip install . -v --user

This may take several minutes to half an hour depending on the speed of your computer.

To test SciPy after installation (highly recommended), execute in Python:

>>> import scipy
>>> scipy.test()

To run the full test suite use:

>>> scipy.test('full')

If you are upgrading from an older SciPy release, please test your code for any deprecation warnings before and after upgrading to avoid surprises:

$ python -Wd -c my_code_that_shouldnt_break.py

Please note that you must have version 1.0 or later of the Pytest test framework installed in order to run the tests. More information about Pytest is available on the website.

You can specify which Fortran compiler to use by using the following install command:

python setup.py config_fc --fcompiler=<Vendor> install

To see a valid list of <Vendor> names, run:

python setup.py config_fc --help-fcompiler

IMPORTANT: It is highly recommended that all libraries that SciPy uses (e.g. BLAS and ATLAS libraries) are built with the same Fortran compiler. In most cases, if you mix compilers, you will not be able to import SciPy at best, and will have crashes and random results at worst.

When installing with python setup.py install or a variation on that, you do not get proper uninstall behavior for an older already installed SciPy version. In many cases that's not a problem, but if it turns out to be an issue, you need to manually uninstall it first (remove from e.g. in /usr/lib/python3.4/site-packages/scipy or $HOME/lib/python3.4/site-packages/scipy).

Alternatively, you can use pip install . --user instead of python setup.py install --user in order to get reliable uninstall behavior. The downside is that pip doesn't show you a build log and doesn't support incremental rebuilds (it copies the whole source tree to a tempdir).

If you experience problems when building/installing/testing SciPy, you can ask help from scipy-user@python.org or scipy-dev@python.org mailing lists. Please include the following information in your message:

NOTE: You can generate some of the following information (items 1-5,7) in one command:

python -c 'from numpy.f2py.diagnose import run; run()'
  1. Platform information:

    python -c 'import os, sys; print(os.name, sys.platform)'
    uname -a
    OS, its distribution name and version information
    etc.
  2. Information about C, C++, Fortran compilers/linkers as reported by the compilers when requesting their version information, e.g., the output of

    gcc -v
    g77 --version
  3. Python version:

    python -c 'import sys; print(sys.version)'
  4. NumPy version:

    python -c 'import numpy; print(numpy.__version__)'
  5. ATLAS version, the locations of atlas and lapack libraries, building information if any. If you have ATLAS version 3.3.6 or newer, then give the output of the last command in

    cd scipy/Lib/linalg
    python setup_atlas_version.py build_ext --inplace --force
    python -c 'import atlas_version'
  1. The output of the following commands

    python INSTALLDIR/numpy/distutils/system_info.py

    where INSTALLDIR is, for example, /usr/lib/python3.4/site-packages/.

  2. Feel free to add any other relevant information. For example, the full output (both stdout and stderr) of the SciPy installation command can be very helpful. Since this output can be rather large, ask before sending it into the mailing list (or better yet, to one of the developers, if asked).

  3. In case of failing to import extension modules, the output of

    ldd /path/to/ext_module.so

    can be useful.