You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

366 lines
11 KiB
Python

5 years ago
"""
Unit test for Linear Programming via Simplex Algorithm.
"""
from __future__ import division, print_function, absolute_import
import numpy as np
from numpy.testing import assert_, assert_allclose
from pytest import raises as assert_raises
from scipy.optimize._linprog_util import _clean_inputs
from copy import deepcopy
def test_aliasing():
c = 1
A_ub = [[1]]
b_ub = [1]
A_eq = [[1]]
b_eq = [1]
bounds = (-np.inf, np.inf)
c_copy = deepcopy(c)
A_ub_copy = deepcopy(A_ub)
b_ub_copy = deepcopy(b_ub)
A_eq_copy = deepcopy(A_eq)
b_eq_copy = deepcopy(b_eq)
bounds_copy = deepcopy(bounds)
_clean_inputs(c, A_ub, b_ub, A_eq, b_eq, bounds)
assert_(c == c_copy, "c modified by _clean_inputs")
assert_(A_ub == A_ub_copy, "A_ub modified by _clean_inputs")
assert_(b_ub == b_ub_copy, "b_ub modified by _clean_inputs")
assert_(A_eq == A_eq_copy, "A_eq modified by _clean_inputs")
assert_(b_eq == b_eq_copy, "b_eq modified by _clean_inputs")
assert_(bounds == bounds_copy, "bounds modified by _clean_inputs")
def test_aliasing2():
c = np.array([1, 1])
A_ub = np.array([[1, 1], [2, 2]])
b_ub = np.array([[1], [1]])
A_eq = np.array([[1, 1]])
b_eq = np.array([1])
bounds = [(-np.inf, np.inf), (None, 1)]
c_copy = c.copy()
A_ub_copy = A_ub.copy()
b_ub_copy = b_ub.copy()
A_eq_copy = A_eq.copy()
b_eq_copy = b_eq.copy()
bounds_copy = deepcopy(bounds)
_clean_inputs(c, A_ub, b_ub, A_eq, b_eq, bounds)
assert_allclose(c, c_copy, err_msg="c modified by _clean_inputs")
assert_allclose(A_ub, A_ub_copy, err_msg="A_ub modified by _clean_inputs")
assert_allclose(b_ub, b_ub_copy, err_msg="b_ub modified by _clean_inputs")
assert_allclose(A_eq, A_eq_copy, err_msg="A_eq modified by _clean_inputs")
assert_allclose(b_eq, b_eq_copy, err_msg="b_eq modified by _clean_inputs")
assert_(bounds == bounds_copy, "bounds modified by _clean_inputs")
def test_missing_inputs():
c = [1, 2]
A_ub = np.array([[1, 1], [2, 2]])
b_ub = np.array([1, 1])
A_eq = np.array([[1, 1], [2, 2]])
b_eq = np.array([1, 1])
assert_raises(TypeError, _clean_inputs)
assert_raises(TypeError, _clean_inputs, c=None)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=A_ub)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=A_ub, b_ub=None)
assert_raises(ValueError, _clean_inputs, c=c, b_ub=b_ub)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=None, b_ub=b_ub)
assert_raises(ValueError, _clean_inputs, c=c, A_eq=A_eq)
assert_raises(ValueError, _clean_inputs, c=c, A_eq=A_eq, b_eq=None)
assert_raises(ValueError, _clean_inputs, c=c, b_eq=b_eq)
assert_raises(ValueError, _clean_inputs, c=c, A_eq=None, b_eq=b_eq)
def test_too_many_dimensions():
cb = [1, 2, 3, 4]
A = np.random.rand(4, 4)
bad2D = [[1, 2], [3, 4]]
bad3D = np.random.rand(4, 4, 4)
assert_raises(ValueError, _clean_inputs, c=bad2D, A_ub=A, b_ub=cb)
assert_raises(ValueError, _clean_inputs, c=cb, A_ub=bad3D, b_ub=cb)
assert_raises(ValueError, _clean_inputs, c=cb, A_ub=A, b_ub=bad2D)
assert_raises(ValueError, _clean_inputs, c=cb, A_eq=bad3D, b_eq=cb)
assert_raises(ValueError, _clean_inputs, c=cb, A_eq=A, b_eq=bad2D)
def test_too_few_dimensions():
bad = np.random.rand(4, 4).ravel()
cb = np.random.rand(4)
assert_raises(ValueError, _clean_inputs, c=cb, A_ub=bad, b_ub=cb)
assert_raises(ValueError, _clean_inputs, c=cb, A_eq=bad, b_eq=cb)
def test_inconsistent_dimensions():
m = 2
n = 4
c = [1, 2, 3, 4]
Agood = np.random.rand(m, n)
Abad = np.random.rand(m, n + 1)
bgood = np.random.rand(m)
bbad = np.random.rand(m + 1)
boundsbad = [(0, 1)] * (n + 1)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=Abad, b_ub=bgood)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=Agood, b_ub=bbad)
assert_raises(ValueError, _clean_inputs, c=c, A_eq=Abad, b_eq=bgood)
assert_raises(ValueError, _clean_inputs, c=c, A_eq=Agood, b_eq=bbad)
assert_raises(ValueError, _clean_inputs, c=c, bounds=boundsbad)
def test_type_errors():
bad = "hello"
c = [1, 2]
A_ub = np.array([[1, 1], [2, 2]])
b_ub = np.array([1, 1])
A_eq = np.array([[1, 1], [2, 2]])
b_eq = np.array([1, 1])
bounds = [(0, 1)]
assert_raises(
TypeError,
_clean_inputs,
c=bad,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=bad,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=bad,
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=bad,
b_eq=b_eq,
bounds=bounds)
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=bad)
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds="hi")
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=["hi"])
assert_raises(
TypeError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=[
("hi")])
assert_raises(TypeError, _clean_inputs, c=c, A_ub=A_ub,
b_ub=b_ub, A_eq=A_eq, b_eq=b_eq, bounds=[(1, "")])
assert_raises(TypeError, _clean_inputs, c=c, A_ub=A_ub,
b_ub=b_ub, A_eq=A_eq, b_eq=b_eq, bounds=[(1, 2), (1, "")])
def test_non_finite_errors():
c = [1, 2]
A_ub = np.array([[1, 1], [2, 2]])
b_ub = np.array([1, 1])
A_eq = np.array([[1, 1], [2, 2]])
b_eq = np.array([1, 1])
bounds = [(0, 1)]
assert_raises(
ValueError, _clean_inputs, c=[0, None], A_ub=A_ub, b_ub=b_ub,
A_eq=A_eq, b_eq=b_eq, bounds=bounds)
assert_raises(
ValueError, _clean_inputs, c=[np.inf, 0], A_ub=A_ub, b_ub=b_ub,
A_eq=A_eq, b_eq=b_eq, bounds=bounds)
assert_raises(
ValueError, _clean_inputs, c=[0, -np.inf], A_ub=A_ub, b_ub=b_ub,
A_eq=A_eq, b_eq=b_eq, bounds=bounds)
assert_raises(
ValueError, _clean_inputs, c=[np.nan, 0], A_ub=A_ub, b_ub=b_ub,
A_eq=A_eq, b_eq=b_eq, bounds=bounds)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=[[1, 2], [None, 1]],
b_ub=b_ub, A_eq=A_eq, b_eq=b_eq, bounds=bounds)
assert_raises(
ValueError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=[
np.inf,
1],
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_raises(ValueError, _clean_inputs, c=c, A_ub=A_ub, b_ub=b_ub, A_eq=[
[1, 2], [1, -np.inf]], b_eq=b_eq, bounds=bounds)
assert_raises(
ValueError,
_clean_inputs,
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=[
1,
np.nan],
bounds=bounds)
def test__clean_inputs1():
c = [1, 2]
A_ub = [[1, 1], [2, 2]]
b_ub = [1, 1]
A_eq = [[1, 1], [2, 2]]
b_eq = [1, 1]
bounds = None
outputs = _clean_inputs(
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_allclose(outputs[0], np.array(c))
assert_allclose(outputs[1], np.array(A_ub))
assert_allclose(outputs[2], np.array(b_ub))
assert_allclose(outputs[3], np.array(A_eq))
assert_allclose(outputs[4], np.array(b_eq))
assert_(outputs[5] == [(0, None)] * 2, "")
assert_(outputs[0].shape == (2,), "")
assert_(outputs[1].shape == (2, 2), "")
assert_(outputs[2].shape == (2,), "")
assert_(outputs[3].shape == (2, 2), "")
assert_(outputs[4].shape == (2,), "")
def test__clean_inputs2():
c = 1
A_ub = [[1]]
b_ub = 1
A_eq = [[1]]
b_eq = 1
bounds = (0, 1)
outputs = _clean_inputs(
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_allclose(outputs[0], np.array(c))
assert_allclose(outputs[1], np.array(A_ub))
assert_allclose(outputs[2], np.array(b_ub))
assert_allclose(outputs[3], np.array(A_eq))
assert_allclose(outputs[4], np.array(b_eq))
assert_(outputs[5] == [(0, 1)], "")
assert_(outputs[0].shape == (1,), "")
assert_(outputs[1].shape == (1, 1), "")
assert_(outputs[2].shape == (1,), "")
assert_(outputs[3].shape == (1, 1), "")
assert_(outputs[4].shape == (1,), "")
def test__clean_inputs3():
c = [[1, 2]]
A_ub = np.random.rand(2, 2)
b_ub = [[1], [2]]
A_eq = np.random.rand(2, 2)
b_eq = [[1], [2]]
bounds = [(0, 1)]
outputs = _clean_inputs(
c=c,
A_ub=A_ub,
b_ub=b_ub,
A_eq=A_eq,
b_eq=b_eq,
bounds=bounds)
assert_allclose(outputs[0], np.array([1, 2]))
assert_allclose(outputs[2], np.array([1, 2]))
assert_allclose(outputs[4], np.array([1, 2]))
assert_(outputs[5] == [(0, 1)] * 2, "")
assert_(outputs[0].shape == (2,), "")
assert_(outputs[2].shape == (2,), "")
assert_(outputs[4].shape == (2,), "")
def test_bad_bounds():
c = [1, 2]
assert_raises(ValueError, _clean_inputs, c=c, bounds=(1, -2))
assert_raises(ValueError, _clean_inputs, c=c, bounds=[(1, -2)])
assert_raises(ValueError, _clean_inputs, c=c, bounds=[(1, -2), (1, 2)])
assert_raises(ValueError, _clean_inputs, c=c, bounds=(1, 2, 2))
assert_raises(ValueError, _clean_inputs, c=c, bounds=[(1, 2, 2)])
assert_raises(ValueError, _clean_inputs, c=c, bounds=[(1, 2), (1, 2, 2)])
assert_raises(ValueError, _clean_inputs, c=c,
bounds=[(1, 2), (1, 2), (1, 2)])
def test_good_bounds():
c = [1, 2]
outputs = _clean_inputs(c=c, bounds=None)
assert_(outputs[5] == [(0, None)] * 2, "")
outputs = _clean_inputs(c=c, bounds=(1, 2))
assert_(outputs[5] == [(1, 2)] * 2, "")
outputs = _clean_inputs(c=c, bounds=[(1, 2)])
assert_(outputs[5] == [(1, 2)] * 2, "")
outputs = _clean_inputs(c=c, bounds=[(1, np.inf)])
assert_(outputs[5] == [(1, None)] * 2, "")
outputs = _clean_inputs(c=c, bounds=[(-np.inf, 1)])
assert_(outputs[5] == [(None, 1)] * 2, "")
outputs = _clean_inputs(c=c, bounds=[(-np.inf, np.inf), (-np.inf, np.inf)])
assert_(outputs[5] == [(None, None)] * 2, "")