self.report_caption(caption)
self.report_caption(self.controls.score.currentText())
if self.score != 0:
self.report_raw(self.get_info_text(short=False))
def set_visual_settings(self, key, value):
self.plot.parameter_setter.set_parameter(key, value)
self.visual_settings[key] = value
def gaussian_smoother(x, y, sigma=1.0):
x = np.asarray(x)
y = np.asarray(y)
gamma = 1. / (2 * sigma ** 2)
a = 1. / (sigma * np.sqrt(2 * np.pi))
if x.shape != y.shape:
raise ValueError
def smoother(xs):
W = a * np.exp(-gamma * ((xs - x) ** 2))
return np.average(y, weights=W)
return np.vectorize(smoother, otypes=[float])
if __name__ == "__main__": # pragma: no cover
WidgetPreview(OWCalibrationPlot).run(results_for_preview())
for i in range(n):
self._set_item(n + 2, i + 2, _sum_item(int(colsum[i]), "t"))
self._set_item(i + 2, n + 2, _sum_item(int(rowsum[i]), "l"))
self._set_item(n + 2, n + 2, _sum_item(int(rowsum.sum())))
def send_report(self):
"""Send report"""
if self.results is not None and self.selected_learner:
self.report_table(
"Confusion matrix for {} (showing {})".format(
self.learners[self.selected_learner[0]],
self.quantities[self.selected_quantity].lower()),
self.tableview)
@classmethod
def migrate_settings(cls, settings, version):
if not version:
# For some period of time the 'selected_learner' property was
# changed from List[int] -> int
# (commit 4e49bb3fd0e11262f3ebf4b1116a91a4b49cc982) and then back
# again (commit 8a492d79a2e17154a0881e24a05843406c8892c0)
if "selected_learner" in settings and \
isinstance(settings["selected_learner"], int):
settings["selected_learner"] = [settings["selected_learner"]]
if __name__ == "__main__": # pragma: no cover
from Orange.widgets.evaluate.utils import results_for_preview
WidgetPreview(OWConfusionMatrix).run(results_for_preview("iris"))
# going through point (0, 1). The point(s) with the minimum
# distance are our result
# y = m * x + 1
# m * x - 1y + 1 = 0
a, b, c = slope, -1, 1
dist = distance_to_line(a, b, c, fpr, tpr)
mindist = np.min(dist)
return np.flatnonzero((dist - mindist) <= tol)
def distance_to_line(a, b, c, x0, y0):
"""
Return the distance to a line ax + by + c = 0
"""
assert a != 0 or b != 0
return np.abs(a * x0 + b * y0 + c) / np.sqrt(a**2 + b**2)
def roc_iso_performance_slope(fp_cost, fn_cost, p):
assert 0 <= p <= 1
if fn_cost * p == 0:
return np.inf
else:
return (fp_cost * (1. - p)) / (fn_cost * p)
if __name__ == "__main__": # pragma: no cover
WidgetPreview(OWROCAnalysis).run(results_for_preview())
caption = report.list_legend(self.classifiers_list_box,
self.selected_classifiers)
self.report_items((("Target class", self.target_cb.currentText()), ))
self.report_plot()
self.report_caption(caption)
def lift_curve_from_results(results, target, clf_idx, subset=slice(0, -1)):
actual = results.actual[subset]
scores = results.probabilities[clf_idx][subset][:, target]
yrate, tpr, thresholds = lift_curve(actual, scores, target)
return yrate, tpr, thresholds
def lift_curve(ytrue, ypred, target=1):
P = np.sum(ytrue == target)
N = ytrue.size - P
if P == 0 or N == 0:
# Undefined TP and FP rate
return np.array([]), np.array([]), np.array([])
fpr, tpr, thresholds = skl_metrics.roc_curve(ytrue, ypred, target)
rpp = fpr * (N / (P + N)) + tpr * (P / (P + N))
return rpp, tpr, thresholds
if __name__ == "__main__": # pragma: no cover
from Orange.widgets.evaluate.utils import results_for_preview
WidgetPreview(OWLiftCurve).run(results_for_preview())
# going through point (0, 1). The point(s) with the minimum
# distance are our result
# y = m * x + 1
# m * x - 1y + 1 = 0
a, b, c = slope, -1, 1
dist = distance_to_line(a, b, c, fpr, tpr)
mindist = numpy.min(dist)
return numpy.flatnonzero((dist - mindist) <= tol)
def distance_to_line(a, b, c, x0, y0):
"""
Return the distance to a line ax + by + c = 0
"""
assert a != 0 or b != 0
return numpy.abs(a * x0 + b * y0 + c) / numpy.sqrt(a ** 2 + b ** 2)
def roc_iso_performance_slope(fp_cost, fn_cost, p):
assert 0 <= p <= 1
if fn_cost * p == 0:
return numpy.inf
else:
return (fp_cost * (1. - p)) / (fn_cost * p)
if __name__ == "__main__": # pragma: no cover
WidgetPreview(OWROCAnalysis).run(results_for_preview())
for i in range(n):
self._set_item(n + 2, i + 2, _sum_item(int(colsum[i]), "t"))
self._set_item(i + 2, n + 2, _sum_item(int(rowsum[i]), "l"))
self._set_item(n + 2, n + 2, _sum_item(int(rowsum.sum())))
def send_report(self):
"""Send report"""
if self.results is not None and self.selected_learner:
self.report_table(
"Confusion matrix for {} (showing {})".
format(self.learners[self.selected_learner[0]],
self.quantities[self.selected_quantity].lower()),
self.tableview)
@classmethod
def migrate_settings(cls, settings, version):
if not version:
# For some period of time the 'selected_learner' property was
# changed from List[int] -> int
# (commit 4e49bb3fd0e11262f3ebf4b1116a91a4b49cc982) and then back
# again (commit 8a492d79a2e17154a0881e24a05843406c8892c0)
if "selected_learner" in settings and \
isinstance(settings["selected_learner"], int):
settings["selected_learner"] = [settings["selected_learner"]]
if __name__ == "__main__": # pragma: no cover
from Orange.widgets.evaluate.utils import results_for_preview
WidgetPreview(OWConfusionMatrix).run(results_for_preview("iris"))
def send_report(self):
if self.results is None:
return
caption = report.list_legend(self.classifiers_list_box,
self.selected_classifiers)
self.report_items((("Target class", self.target_cb.currentText()),))
self.report_plot()
self.report_caption(caption)
def gaussian_smoother(x, y, sigma=1.0):
x = np.asarray(x)
y = np.asarray(y)
gamma = 1. / (2 * sigma ** 2)
a = 1. / (sigma * np.sqrt(2 * np.pi))
if x.shape != y.shape:
raise ValueError
def smoother(xs):
W = a * np.exp(-gamma * ((xs - x) ** 2))
return np.average(y, weights=W)
return np.vectorize(smoother, otypes=[np.float])
if __name__ == "__main__": # pragma: no cover
WidgetPreview(OWCalibrationPlot).run(results_for_preview())
caption = report.list_legend(self.classifiers_list_box,
self.selected_classifiers)
self.report_items((("Target class", self.target_cb.currentText()),))
self.report_plot()
self.report_caption(caption)
def lift_curve_from_results(results, target, clf_idx, subset=slice(0, -1)):
actual = results.actual[subset]
scores = results.probabilities[clf_idx][subset][:, target]
yrate, tpr, thresholds = lift_curve(actual, scores, target)
return yrate, tpr, thresholds
def lift_curve(ytrue, ypred, target=1):
P = np.sum(ytrue == target)
N = ytrue.size - P
if P == 0 or N == 0:
# Undefined TP and FP rate
return np.array([]), np.array([]), np.array([])
fpr, tpr, thresholds = skl_metrics.roc_curve(ytrue, ypred, target)
rpp = fpr * (N / (P + N)) + tpr * (P / (P + N))
return rpp, tpr, thresholds
if __name__ == "__main__": # pragma: no cover
from Orange.widgets.evaluate.utils import results_for_preview
WidgetPreview(OWLiftCurve).run(results_for_preview())