Something went wrong. Try again.
What happens if I push all of nixpkgs
Something went wrong. Try again.
1.8 kB · 90 lines
Nix
12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091{ lib, buildPythonPackage, fetchFromGitHub,
# build-system setuptools,
# dependencies absl-py, jax, matplotlib, numpy, scipy,
# tests cvxpy, optax, pytest-xdist, pytestCheckHook, scikit-learn,}:
buildPythonPackage rec { pname = "jaxopt"; version = "0.8.5"; pyproject = true;
src = fetchFromGitHub { owner = "google"; repo = "jaxopt"; tag = "jaxopt-v${version}"; hash = "sha256-vPXrs8J81O+27w9P/fEFr7w4xClKb8T0IASD+iNhztQ="; };
build-system = [ setuptools ];
dependencies = [ absl-py jax matplotlib numpy scipy ];
nativeCheckInputs = [ cvxpy optax pytest-xdist pytestCheckHook scikit-learn ];
pythonImportsCheck = [ "jaxopt" "jaxopt.implicit_diff" "jaxopt.linear_solve" "jaxopt.loss" "jaxopt.tree_util" ];
disabledTests = [ # https://github.com/google/jaxopt/issues/592 "test_solve_sparse"
# https://github.com/google/jaxopt/issues/593 # Makes the test suite crash "test_dtype_consistency"
# AssertionError: Not equal to tolerance rtol=1e-06, atol=1e-06 # https://github.com/google/jaxopt/issues/618 "test_binary_logit_log_likelihood"
# AssertionError (flaky numerical tests) "test_Rosenbrock2" "test_Rosenbrock5" "test_gradient1" "test_inv_hessian_product_pytree3" "test_logreg_with_intercept_manual_loop3" "test_multiclass_logreg6" ];
meta = { homepage = "https://jaxopt.github.io"; description = "Hardware accelerated, batchable and differentiable optimizers in JAX"; changelog = "https://github.com/google/jaxopt/releases/tag/jaxopt-v${version}"; license = lib.licenses.asl20; maintainers = with lib.maintainers; [ bcdarwin ]; };}