movefasta/nixpkgs
0
0
Fork 0
mirror of https://github.com/NixOS/nixpkgs.git synced 2025-07-13 13:40:28 +03:00
Code Activity

pythonPackages.tensorflow: 1.11 -> 1.13.1

Browse source
This commit is contained in:
Jean-Philippe Bernardy 2019-02-28 11:59:17 +01:00 • committed by Frederik Rietdijk
parent d1b7ada0ba
commit 1c350dc99e
6 changed files with 41 additions and 75 deletions
Download patch file Download diff file Expand all files Collapse all files

8
pkgs/development/python-modules/tensorflow/bin.nix
View file

@ -14,6 +14,7 @@
, mock
, backports_weakref
, enum34
, tensorflow-estimator
, tensorflow-tensorboard
, cudaSupport ? false
, cudatoolkit ? null
@ -41,7 +42,7 @@ let
in buildPythonPackage rec {
pname = "tensorflow";
version = "1.11.0";
version = "1.13.1";
format = "wheel";
src = let
@ -50,10 +51,10 @@ in buildPythonPackage rec {
platform = if stdenv.isDarwin then "mac" else "linux";
unit = if cudaSupport then "gpu" else "cpu";
key = "${platform}_py_${pyver}_${unit}";
dls = import ./tf1.11.0-hashes.nix;
dls = import (./. + "/tf${version}-hashes.nix");
in fetchurl dls.${key};
propagatedBuildInputs = [ protobuf numpy termcolor grpcio six astor absl-py gast tensorflow-tensorboard keras-applications keras-preprocessing ]
propagatedBuildInputs = [ protobuf numpy termcolor grpcio six astor absl-py gast tensorflow-estimator tensorflow-tensorboard keras-applications keras-preprocessing ]
++ lib.optional (!isPy3k) mock;
# Upstream has a pip hack that results in bin/tensorboard being in both tensorflow
@ -64,7 +65,6 @@ in buildPythonPackage rec {
rm $out/bin/tensorboard
'';
installFlags = "--no-dependencies"; # tensorflow wants setuptools 39, can't allow that.
# Note that we need to run *after* the fixup phase because the
# libraries are loaded at runtime. If we run in preFixup then
# patchelf --shrink-rpath will remove the cuda libraries.