Tensorflow: Difference between revisions

From NixOS Wiki
imported>Samuela
No edit summary
Klinger (talk | contribs)
m added Category:Python, link, description
 
(One intermediate revision by one other user not shown)
Line 1: Line 1:
[https://www.tensorflow.org/ TensorFlow] is a free and open-source software library for machine learning and artificial intelligence.
= Tensorflow =
= Tensorflow =
There several possible ways to install tensorflow.


== Nix-native packages (strongly recommended) ==
=== Quick shell <ref name=":0">https://discourse.nixos.org/t/cuda-tensorflow-my-setup-is-really-hacky-would-appreciate-help-unhackying-it/43912/2</ref> ===
Use <code>python3Packages.tensorflow</code> or <code>python3Packages.tensorflowWithCuda</code>.
nix-shell --arg config '{ cudaSupport = true; allowUnfree = true; }' -p 'python3.withPackages (ps: [ ps.tensorflow ]<code>)'</code>
This starts a shell with tensorflow and it has GPU support.


'''Cache''': Using the [https://app.cachix.org/cache/cuda-maintainers#pull cuda-maintainers cache] is recommended! It will save you valuable time and electrons. Getting set up should be as simple as <code>cachix use cuda-maintainers</code>. See the [[CUDA]] wiki page for more info.
=== Shell config ===
A basic shell config would look something like this:
<code># shell.nix
{ pkgs ? import <nixpkgs> {}, config ? {} }:
let
<nowiki> </nowiki> pythonPackages = pkgs.python3.withPackages (ps: [ ps.tensorflow ]);
in
pkgs.mkShell {
<nowiki> </nowiki> name = "tf";
<nowiki> </nowiki> buildInputs = [
<nowiki> </nowiki>   (pythonPackages)
<nowiki> </nowiki> ];
<nowiki> </nowiki> shellHook = <nowiki>''</nowiki>
<nowiki> </nowiki>   export PYTHONPATH="${pythonPackages}:${PYTHONPATH:-}"
<nowiki> </nowiki> <nowiki>''</nowiki>;
}</code>
To make this configuration work, edit <code>~/.config/nxpkgs/config.nix</code>:
<code># ~/.config/nxpkgs/config.nix
{ cudaSupport = true; allowUnfree = true;  }</code>


== pip install in a nix-shell ==
== References ==
Nixpkgs provides multiple versions, however, it is often desirable to be able to install the latest nightly from pip. This can accomplished in the following ways:
<ref name=":0" />


* By making a nix-shell
[[Category:Python]]
 
<syntaxhighlight lang="nix">
with import <nixpkgs> {};
mkShell {
  name = "tensorflow-cuda-shell";
  buildInputs = with python3.pkgs; [
    pip
    numpy
    setuptools
  ];
  shellHook = ''
    export LD_LIBRARY_PATH=${pkgs.stdenv.cc.cc.lib}/lib:${pkgs.cudatoolkit_10_1}/lib:${pkgs.cudnn_cudatoolkit_10_1}/lib:${pkgs.cudatoolkit_10_1.lib}/lib:$LD_LIBRARY_PATH
    alias pip="PIP_PREFIX='$(pwd)/_build/pip_packages' TMPDIR='$HOME' \pip"
    export PYTHONPATH="$(pwd)/_build/pip_packages/lib/python3.7/site-packages:$PYTHONPATH"
    export PATH="$(pwd)/_build/pip_packages/bin:$PATH"
    unset SOURCE_DATE_EPOCH
  '';
}
</syntaxhighlight>
 
Within this shell, <code>pip install tf-nightly</code> should work and provide GPU support. The cuda toolkit version (and the version of Python) can be changed to correspond with the matching tensorflow version.
 
Note: On NixOS 20.03 and above LD_LIBRARY_PATH no longer contains /run/opengl-driver/lib:/run/opengl-driver-32/lib by default, preventing tensorflow from discovering the Cuda libraries. This can be solved by manually (or using nixGL[https://github.com/guibou/nixGL] ) appending your LD_LIBRARY_PATH in the shellHook, or by reverting to the pre-20.03 behavior by setting setLdLibraryPath to true in your hardware opengl configuration under configuration.nix.
 
hardware.opengl.setLdLibraryPath = true;
 
== See also ==
 
* [https://discourse.nixos.org/t/installing-python-modules-from-pip-buildfhsuserenv-for-python-tutorial/5704/13]

Latest revision as of 18:07, 25 August 2024

TensorFlow is a free and open-source software library for machine learning and artificial intelligence.

Tensorflow

Quick shell [1]

nix-shell --arg config '{ cudaSupport = true; allowUnfree = true; }' -p 'python3.withPackages (ps: [ ps.tensorflow ])'

This starts a shell with tensorflow and it has GPU support.

Shell config

A basic shell config would look something like this:

# shell.nix
{ pkgs ? import <nixpkgs> {}, config ? {} }:
let
  pythonPackages = pkgs.python3.withPackages (ps: [ ps.tensorflow ]);
in
pkgs.mkShell {
  name = "tf";
  buildInputs = [
    (pythonPackages)
  ];
  shellHook = ''
    export PYTHONPATH="${pythonPackages}:${PYTHONPATH:-}"
  '';
}

To make this configuration work, edit ~/.config/nxpkgs/config.nix:

# ~/.config/nxpkgs/config.nix
{ cudaSupport = true; allowUnfree = true;  }

References

[1]