def test_feature_importance_errors_non_dataframe():
x = np.random.random(100)
y = np.random.random(100)
z = ["pos", "neg"] * 50
df = pd.DataFrame(list(zip(x, y, z)))
df.columns = ["x", "y", "Metadata_compound"]
with pytest.raises(ValueError):
feature_selection.feature_importance(x, "pos", "neg", compound_col="incorrect")
def test_feature_importance_returns_all_feature_columns():
x, y = make_classification(n_samples=100, n_features=10, n_informative=2)
x = pd.DataFrame(x)
x.columns = ["x" + str(i) for i in range(1, 11)]
x["Metadata_compound"] = ["pos", "neg"] * 50
out = feature_selection.feature_importance(
data=x, neg_cmpd="neg", pos_cmpd="pos", compound_col="Metadata_compound"
)
assert len(list(out)) == 10
def test_feature_importance_returns_colnames():
x, y = make_classification(n_samples=100, n_features=10, n_informative=2)
x = pd.DataFrame(x)
x.columns = ["x" + str(i) for i in range(1, 11)]
x["Metadata_compound"] = ["pos", "neg"] * 50
out = feature_selection.feature_importance(
data=x, neg_cmpd="neg", pos_cmpd="pos", compound_col="Metadata_compound"
)
f_names, importances = list(zip(*out))
feature_col_names = utils.get_featuredata(x)
assert list(f_names) == list(feature_col_names)
def test_feature_importance_sort():
x, y = make_classification(n_samples=100, n_features=10, n_informative=2)
x = pd.DataFrame(x)
x.columns = ["x" + str(i) for i in range(1, 11)]
x["Metadata_compound"] = ["pos", "neg"] * 50
out = feature_selection.feature_importance(
data=x,
neg_cmpd="neg",
pos_cmpd="pos",
compound_col="Metadata_compound",
sort=True)
f_names, importances = list(zip(*out))
sorted_importances = sorted(list(importances), reverse=True)
assert sorted_importances == list(importances)
def test_feature_importance_sort():
x, y = make_classification(n_samples=100, n_features=10, n_informative=2)
x = pd.DataFrame(x)
x.columns = ["x" + str(i) for i in range(1, 11)]
x["Metadata_compound"] = ["pos", "neg"] * 50
out = feature_selection.feature_importance(
data=x,
neg_cmpd="neg",
pos_cmpd="pos",
compound_col="Metadata_compound",
sort=True,
)
f_names, importances = list(zip(*out))
sorted_importances = sorted(list(importances), reverse=True)
assert sorted_importances == list(importances)
