def test_weights_input():
tp = TOPSIS(dec_mat_2, cost_ben, weights=[1, 2, 3, 4])
assert np.isclose(tp.weights.sum(), 1.0)
with pytest.raises(ValueError):
_ = TOPSIS(dec_mat_2, cost_ben, weights=[1, 2, 3])
with pytest.raises(ValueError):
_ = TOPSIS(dec_mat_2, cost_ben, weights=10)
def test_input_from_csv_file_alternative():
tp = TOPSIS("dec_mat_2.csv",
cost_ben,
weights=weights,
crit_col_names=criteria,
alt_col_name="alternative")
assert is_equal(list(tp.alternatives), alternatives)
def test_input_from_csv_file_normal_flow():
tp = TOPSIS("dec_mat_2.csv",
cost_ben,
weights=weights,
crit_col_names=criteria,
alt_col_name="alternative")
assert np.allclose(tp.matrix_d, dec_norm_w)
dec_mat_2 = [[15, 6, 25000, 7], [12, 7, 35000, 7], [10, 9, 55000, 8]]
alternatives = ["Alt 1", "Alt 2", "Alt 3"]
criteria = ["criterion 1", "criterion 2", "criterion 3", "criterion 4"]
weights = [0.3, 0.05, 0.6, 0.05]
cost_ben = ["c", "b", "c", "b"]
########################################################################################################################
# Approach 1: using the csv file in "../test/dec_mat_2.csv"
########################################################################################################################
print("-" * 50)
print("- Approach 1:")
print("-" * 50)
tp = TOPSIS("../test/dec_mat_2.csv",
cost_ben,
weights=weights,
alt_col_name="alternative",
crit_col_names=criteria)
tp.get_closeness_coefficient(verbose=True)
tp.plot_ranking()
print("-" * 50)
print("")
########################################################################################################################
# Approach 2: using the matrix as a list of list (but it could be a numpy array as well
########################################################################################################################
print("-" * 50)
print("- Approach 2:")
print("-" * 50)
tp = TOPSIS(dec_mat_2, cost_ben, weights=weights)
tp.get_closeness_coefficient(verbose=True)
def test_ideal_solutions():
tp = TOPSIS(dec_mat_2, cost_ben, weights=weights)
tp.get_ideal_solutions()
assert np.allclose(tp.ideal_neg, ideal_neg)
assert np.allclose(ideal_pos, ideal_pos)
def test_weights_none():
tp = TOPSIS(dec_mat_2, cost_ben)
assert tp.n_crit == len(tp.weights)
assert np.isclose(tp.weights.sum(), 1.0)
def test_num_crit_alter():
tp = TOPSIS(dec_mat_2, cost_ben)
assert tp.n_alt == 3
assert tp.n_crit == 4
def test_input_invalid():
with pytest.raises(ValueError):
tp = TOPSIS(10, cost_ben)
def test_input_from_numpy():
tp = TOPSIS(np.array(dec_mat_2), cost_ben, weights=weights)
assert np.allclose(tp.matrix_d, dec_norm_w)
def test_input_from_list():
tp = TOPSIS(dec_mat_2, cost_ben, weights=weights)
assert np.allclose(tp.matrix_d, dec_norm_w)
def test_cost_ben_str():
tp = TOPSIS("dec_mat_2.csv",
"cost",
crit_col_names=criteria,
alt_col_name="alternative")
assert is_equal(list(tp.cost_ben), ["cost"] * 4)
def test_input_from_csv_file_criteria():
tp = TOPSIS("dec_mat_2.csv",
cost_ben,
crit_col_names=criteria,
alt_col_name="alternative")
assert is_equal(list(tp.criteria), criteria)
def test_clos_coefficient():
tp = TOPSIS(dec_mat_2, cost_ben, weights=weights)
tp.get_closeness_coefficient()
assert np.allclose(tp.clos_coefficient, clos_coefficient)
def test_ideal_distances():
tp = TOPSIS(dec_mat_2, cost_ben, weights=weights)
tp.get_distance_to_ideal()
assert np.allclose(tp.dist_neg, dist_neg)
assert np.allclose(tp.dist_pos, dist_pos)