alioth/nixpkgs
3
0
Fork 0
forked from mirrors/nixpkgs
Code Releases Activity
nixpkgs/pkgs/development/python-modules/sentencepiece/default.nix

19 lines
380 B
Nix
Raw Normal View History

pythonPackages.sentencepiece: init at 0.1.84
2019-12-07 03:26:02 +00:00
{ buildPythonPackage
, stdenv
, sentencepiece
, pkgconfig
}:
buildPythonPackage rec {
pname = "sentencepiece";
sentencepiece: split into multiple outputs, optional gperftools I am using sentencepiece in a downstream application where I want to minimize the resulting closures. This commit makes changes to make sentencepiece a leaner dependency: - Split the outputs, so that the binaries/headers do not end up in the transitive closure in a library dependency. - Add the `withGPerfTools` option, which is enabled by default, to make it possible to disable the gperftools dependency. According to the sentencepiece README, this dependency gives a 10-40% performance improvement. But in many cases this is overshadowed by the neural networks that use piece identifiers as input anyway.
2020-02-25 15:06:24 +00:00
inherit (sentencepiece) version src;
pythonPackages.sentencepiece: init at 0.1.84
2019-12-07 03:26:02 +00:00
nativeBuildInputs = [ pkgconfig ];
sentencepiece: split into multiple outputs, optional gperftools I am using sentencepiece in a downstream application where I want to minimize the resulting closures. This commit makes changes to make sentencepiece a leaner dependency: - Split the outputs, so that the binaries/headers do not end up in the transitive closure in a library dependency. - Add the `withGPerfTools` option, which is enabled by default, to make it possible to disable the gperftools dependency. According to the sentencepiece README, this dependency gives a 10-40% performance improvement. But in many cases this is overshadowed by the neural networks that use piece identifiers as input anyway.
2020-02-25 15:06:24 +00:00
buildInputs = [ sentencepiece.dev ];
pythonPackages.sentencepiece: init at 0.1.84
2019-12-07 03:26:02 +00:00
sourceRoot = "source/python";
sentencepiece: split into multiple outputs, optional gperftools I am using sentencepiece in a downstream application where I want to minimize the resulting closures. This commit makes changes to make sentencepiece a leaner dependency: - Split the outputs, so that the binaries/headers do not end up in the transitive closure in a library dependency. - Add the `withGPerfTools` option, which is enabled by default, to make it possible to disable the gperftools dependency. According to the sentencepiece README, this dependency gives a 10-40% performance improvement. But in many cases this is overshadowed by the neural networks that use piece identifiers as input anyway.
2020-02-25 15:06:24 +00:00
# sentencepiece installs 'bin' output.
meta = builtins.removeAttrs sentencepiece.meta [ "outputsToInstall" ];
pythonPackages.sentencepiece: init at 0.1.84
2019-12-07 03:26:02 +00:00
}
Copy permalink