depot/third_party/nixpkgs/pkgs/development/python-modules/tbats/default.nix
Default email 5e7c2d6cef Project import generated by Copybara.
GitOrigin-RevId: f99e5f03cc0aa231ab5950a15ed02afec45ed51a
2023-10-09 21:29:22 +02:00

56 lines
1.2 KiB
Nix

{ lib
, buildPythonPackage
, fetchFromGitHub
, setuptools
, numpy
, pmdarima
, scikit-learn
, scipy
, pytestCheckHook
}:
buildPythonPackage rec {
pname = "tbats";
version = "1.1.3";
pyproject = true;
src = fetchFromGitHub {
owner = "intive-DataScience";
repo = "tbats";
rev = version;
hash = "sha256-f6QqDq/ffbnFBZRAT6KQRlqvZZSE+Pff2/o+htVabZI=";
};
nativeBuildInputs = [
setuptools
];
propagatedBuildInputs = [
numpy
pmdarima
scikit-learn
scipy
];
nativeCheckInputs = [ pytestCheckHook ];
pytestFlagsArray = [
# test_R folder is just for comparison of results with R lib
# we need only test folder
"test/"
# several tests has same name, so we use --deselect instead of disableTests
# Test execution is too long > 15 min
"--deselect=test/tbats/TBATS_test.py::TestTBATS::test_fit_predict_trigonometric_seasonal"
];
pythonImportsCheck = [ "tbats" ];
meta = with lib; {
description = "BATS and TBATS forecasting methods";
homepage = "https://github.com/intive-DataScience/tbats";
changelog = "https://github.com/intive-DataScience/tbats/releases/tag/${src.rev}";
license = licenses.mit;
maintainers = with maintainers; [ mbalatsko ];
};
}