Tensorflow - Revision history

Klinger: added Category:Python, link, description

2024-08-25T18:07:56Z

added Category:Python, link, description

← Older revision		Revision as of 18:07, 25 August 2024
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 =

Line 27:		Line 28:
	== References ==		== References ==
	<ref name=":0" />		<ref name=":0" />

			[[Category:Python]]

Sv-cheats-1: Updated wiki page on tensorflow. Old configuration didn't work anymore, so I provided one that worked for me

2024-05-02T11:24:10Z

Updated wiki page on tensorflow. Old configuration didn't work anymore, so I provided one that worked for me

@@ Line 1: / Line 1: @@
 = 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>
-'''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 ===
-== 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" />

imported>Samuela at 19:28, 31 March 2022

2022-03-31T19:28:47Z

← Older revision		Revision as of 19:28, 31 March 2022
Line 1:		Line 1:
	== Tensorflow ==		= Tensorflow =
	There several possible ways to install tensorflow.		There several possible ways to install tensorflow.

	=== Nix-native packages (recommended) ===		== Nix-native packages (strongly recommended) ==
	Use <code>python3Packages.tensorflow</code> or <code>python3Packages.tensorflowWithCuda</code>.		Use <code>python3Packages.tensorflow</code> or <code>python3Packages.tensorflowWithCuda</code>.

	=== pip install in a nix-shell ===		'''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.

			== pip install in a nix-shell ==
	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:		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:

Line 32:		Line 34:

	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.		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;		hardware.opengl.setLdLibraryPath = true;



	== See also ==		== See also ==

	* [https://discourse.nixos.org/t/installing-python-modules-from-pip-buildfhsuserenv-for-python-tutorial/5704/13]		* [https://discourse.nixos.org/t/installing-python-modules-from-pip-buildfhsuserenv-for-python-tutorial/5704/13]

imported>Samuela at 02:43, 1 February 2022

2022-02-01T02:43:03Z

← Older revision		Revision as of 02:43, 1 February 2022
Line 1:		Line 1:
	== Tensorflow ==		== Tensorflow ==
			There several possible ways to install tensorflow.

	~~There several possible ways to install~~ tensorflow. 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:		=== Nix-native packages (recommended) ===
			Use <code>python3Packages.tensorflow</code> or <code>python3Packages.tensorflowWithCuda</code>.

			=== pip install in a nix-shell ===
			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:

	* By making a nix-shell		* By making a nix-shell
Line 24:		Line 29:
	</syntaxhighlight>		</syntaxhighlight>

	Within this shell, pip install tf-nightly 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.		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.		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.

imported>Biggs: Note about changing python version to be up to date

2021-10-18T11:28:03Z

Note about changing python version to be up to date

← Older revision		Revision as of 11:28, 18 October 2021
Line 24:		Line 24:
	</syntaxhighlight>		</syntaxhighlight>

	Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version (and the version of ~~tensorflow~~) can be changed to correspond with the matching tensorflow version.		Within this shell, pip install tf-nightly 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.		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.

imported>Biggs at 11:27, 18 October 2021

2021-10-18T11:27:16Z

← Older revision		Revision as of 11:27, 18 October 2021
Line 24:		Line 24:
	</syntaxhighlight>		</syntaxhighlight>

	Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version can be changed to correspond with the matching tensorflow version.		Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version (and the version of tensorflow) 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.		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.

imported>Mjlbach at 07:23, 12 May 2020

2020-05-12T07:23:35Z

← Older revision		Revision as of 07:23, 12 May 2020
Line 26:		Line 26:
	Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version can be changed to correspond with the matching tensorflow version.		Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version 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), or by reverting to the pre-20.03 behavior by setting setLdLibraryPath to true in your hardware opengl configuration under configuration.nix.		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.

imported>Mjlbach at 07:21, 12 May 2020

2020-05-12T07:21:04Z

← Older revision		Revision as of 07:21, 12 May 2020
Line 25:		Line 25:

	Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version can be changed to correspond with the matching tensorflow version.		Within this shell, pip install tf-nightly should work and provide GPU support. The cuda toolkit version 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), 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 ==		== See also ==

	* [https://discourse.nixos.org/t/installing-python-modules-from-pip-buildfhsuserenv-for-python-tutorial/5704/13]		* [https://discourse.nixos.org/t/installing-python-modules-from-pip-buildfhsuserenv-for-python-tutorial/5704/13]

imported>Mjlbach at 02:04, 23 February 2020

2020-02-23T02:04:45Z

← Older revision		Revision as of 02:04, 23 February 2020
Line 15:		Line 15:
	];		];
	shellHook = ''		shellHook = ''
	export LD_LIBRARY_PATH=~~${pkgs.linuxPackages.nvidia_x11}/lib:~~${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		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='$HOME' \pip"		alias pip="PIP_PREFIX='$(pwd)/_build/pip_packages' TMPDIR='$HOME' \pip"
	export PYTHONPATH="$(pwd)/_build/pip_packages/lib/python3.7/site-packages:$PYTHONPATH"		export PYTHONPATH="$(pwd)/_build/pip_packages/lib/python3.7/site-packages:$PYTHONPATH"

imported>Mjlbach at 02:37, 14 February 2020

2020-02-14T02:37:47Z

← Older revision		Revision as of 02:37, 14 February 2020
Line 1:		Line 1:
	== Tensorflow ==		== Tensorflow ==

	There ~~some~~ possible ways to install tensorflow. Nixpkgs provides ~~multiiple~~ versions, however, it is often desirable to be able to install the latest nightly from pip. This can accomplished in the following ways:		There several possible ways to install tensorflow. 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:

	* By making a nix-shell		* By making a nix-shell