diff --git a/pkgs/development/libraries/sentencepiece/default.nix b/pkgs/development/libraries/sentencepiece/default.nix new file mode 100644 index 000000000000..556f77ef8b51 --- /dev/null +++ b/pkgs/development/libraries/sentencepiece/default.nix @@ -0,0 +1,31 @@ +{ config +, fetchFromGitHub +, stdenv +, lib +, cmake +, gperftools +}: + +stdenv.mkDerivation rec { + pname = "sentencepiece"; + version = "0.1.84"; + + src = fetchFromGitHub { + owner = "google"; + repo = pname; + rev = "v${version}"; + sha256 = "144y25nj4rwxmgvzqbr7al9fjwh3539ssjswvzrx4gsgfk62lsm0"; + }; + + enableParallelBuilding = true; + + nativeBuildInputs = [ cmake gperftools ]; + + meta = with stdenv.lib; { + homepage = https://github.com/google/sentencepiece; + description = "Unsupervised text tokenizer for Neural Network-based text generation"; + license = licenses.asl20; + platforms = [ "x86_64-linux" ]; + maintainers = with maintainers; [ pashashocky ]; + }; +} diff --git a/pkgs/top-level/all-packages.nix b/pkgs/top-level/all-packages.nix index e11f942c6fb3..d71a039199cb 100644 --- a/pkgs/top-level/all-packages.nix +++ b/pkgs/top-level/all-packages.nix @@ -25611,4 +25611,6 @@ in gortr = callPackage ../servers/gortr {}; + sentencepiece = callPackage ../development/libraries/sentencepiece {}; + }