def test_plot_det_curve(pyplot, response_method, data_binary,
with_sample_weight, with_strings):
X, y = data_binary
pos_label = None
if with_strings:
y = np.array(["c", "b"])[y]
pos_label = "c"
if with_sample_weight:
rng = np.random.RandomState(42)
sample_weight = rng.randint(1, 4, size=(X.shape[0]))
else:
sample_weight = None
lr = LogisticRegression()
lr.fit(X, y)
viz = plot_det_curve(
lr,
X,
y,
alpha=0.8,
sample_weight=sample_weight,
)
y_pred = getattr(lr, response_method)(X)
if y_pred.ndim == 2:
y_pred = y_pred[:, 1]
fpr, fnr, _ = det_curve(
y,
y_pred,
sample_weight=sample_weight,
pos_label=pos_label,
)
assert_allclose(viz.fpr, fpr)
assert_allclose(viz.fnr, fnr)
assert viz.estimator_name == "LogisticRegression"
# cannot fail thanks to pyplot fixture
import matplotlib as mpl # noqal
assert isinstance(viz.line_, mpl.lines.Line2D)
assert viz.line_.get_alpha() == 0.8
assert isinstance(viz.ax_, mpl.axes.Axes)
assert isinstance(viz.figure_, mpl.figure.Figure)
assert viz.line_.get_label() == "LogisticRegression"
expected_pos_label = 1 if pos_label is None else pos_label
expected_ylabel = (
f"False Negative Rate (Positive label: {expected_pos_label})")
expected_xlabel = (
f"False Positive Rate (Positive label: {expected_pos_label})")
assert viz.ax_.get_ylabel() == expected_ylabel
assert viz.ax_.get_xlabel() == expected_xlabel
def no_step_dump(self) -> int:
from sklearn import metrics
fpr, fnr, roc_thresholds = metrics.det_curve(y_true=self.val[0],
y_score=self.val[1],
**self._dump_kwargs)
det = {
"det": [{
"fpr": fp,
"fnr": fn,
"threshold": t
} for fp, fn, t in zip(fpr, fnr, roc_thresholds)]
}
self.write_json(det, self.output_path)
def test_det_curve_display(pyplot, constructor_name, response_method,
with_sample_weight, with_strings):
X, y = load_iris(return_X_y=True)
# Binarize the data with only the two first classes
X, y = X[y < 2], y[y < 2]
pos_label = None
if with_strings:
y = np.array(["c", "b"])[y]
pos_label = "c"
if with_sample_weight:
rng = np.random.RandomState(42)
sample_weight = rng.randint(1, 4, size=(X.shape[0]))
else:
sample_weight = None
lr = LogisticRegression()
lr.fit(X, y)
y_pred = getattr(lr, response_method)(X)
if y_pred.ndim == 2:
y_pred = y_pred[:, 1]
# safe guard for the binary if/else construction
assert constructor_name in ("from_estimator", "from_predictions")
common_kwargs = {
"name": lr.__class__.__name__,
"alpha": 0.8,
"sample_weight": sample_weight,
"pos_label": pos_label,
}
if constructor_name == "from_estimator":
disp = DetCurveDisplay.from_estimator(lr, X, y, **common_kwargs)
else:
disp = DetCurveDisplay.from_predictions(y, y_pred, **common_kwargs)
fpr, fnr, _ = det_curve(
y,
y_pred,
sample_weight=sample_weight,
pos_label=pos_label,
)
assert_allclose(disp.fpr, fpr)
assert_allclose(disp.fnr, fnr)
assert disp.estimator_name == "LogisticRegression"
# cannot fail thanks to pyplot fixture
import matplotlib as mpl # noqal
assert isinstance(disp.line_, mpl.lines.Line2D)
assert disp.line_.get_alpha() == 0.8
assert isinstance(disp.ax_, mpl.axes.Axes)
assert isinstance(disp.figure_, mpl.figure.Figure)
assert disp.line_.get_label() == "LogisticRegression"
expected_pos_label = 1 if pos_label is None else pos_label
expected_ylabel = f"False Negative Rate (Positive label: {expected_pos_label})"
expected_xlabel = f"False Positive Rate (Positive label: {expected_pos_label})"
assert disp.ax_.get_ylabel() == expected_ylabel
assert disp.ax_.get_xlabel() == expected_xlabel