def test_cmap(self):
np.random.seed(0)
clf = LogisticRegression()
clf.fit(self.X, self.y)
probas = clf.predict_proba(self.X)
plot_roc_curve(self.y, probas, cmap='nipy_spectral')
plot_roc_curve(self.y, probas, cmap=plt.cm.nipy_spectral)
def test_curve_diffs(self):
np.random.seed(0)
clf = LogisticRegression()
clf.fit(self.X, self.y)
probas = clf.predict_proba(self.X)
ax_macro = plot_roc_curve(self.y, probas, curves='macro')
ax_micro = plot_roc_curve(self.y, probas, curves='micro')
ax_class = plot_roc_curve(self.y, probas, curves='each_class')
self.assertNotEqual(ax_macro, ax_micro, ax_class)
def test_ax(self):
np.random.seed(0)
clf = LogisticRegression()
clf.fit(self.X, self.y)
probas = clf.predict_proba(self.X)
fig, ax = plt.subplots(1, 1)
out_ax = plot_roc_curve(self.y, probas)
assert ax is not out_ax
out_ax = plot_roc_curve(self.y, probas, ax=ax)
assert ax is out_ax
def plot_analysis(combine,
test_name,
y_true,
y_pred,
y_proba,
labels,
verbose,
library,
save=True,
show=True,
sessionid="testing",
prefix=""):
met_index = 0
plt.rcParams.update({'font.size': 14})
# TODO: Find a way to do this better
pltmetrics.plot_confusion_matrix(y_true, y_pred)
if not combine:
#plt.gcf().set_size_inches(3.65,3.65)
save_show(plt, library + "/" + prefix, sessionid, "confusion_matrix",
show, save, False, True, True, False)
else:
plt.subplot(2, 4, met_index + 1)
met_index += 1
plt.rcParams.update({'font.size': 12})
pltmetrics.plot_roc_curve(y_true, y_proba)
for text in plt.gca().legend_.get_texts():
text.set_text(text.get_text().replace("ROC curve of class", "class"))
text.set_text(text.get_text().replace("area =", "AUC: "))
text.set_text(text.get_text().replace("micro-average ROC curve",
"micro-avg"))
text.set_text(text.get_text().replace("macro-average ROC curve",
"macro-avg"))
if not combine:
#plt.gcf().set_size_inches(3.65,3.65)
save_show(plt, library + "/" + prefix, sessionid, "roc_curves", show,
save, False, True, True, False)
else:
plt.subplot(2, 4, met_index + 1)
met_index += 1
if len(labels) < 3:
pltmetrics.plot_ks_statistic(y_true, y_proba)
if not combine:
#plt.gcf().set_size_inches(3.65,3.65)
save_show(plt, library + "/" + prefix, sessionid, "ks_statistics",
show, save, False, True, True, False)
else:
plt.subplot(2, 4, met_index + 1)
met_index += 1
pltmetrics.plot_precision_recall_curve(y_true, y_proba)
for text in plt.gca().legend_.get_texts():
text.set_text(text.get_text().replace(
"Precision-recall curve of class", "class"))
text.set_text(text.get_text().replace("area =", "AUC: "))
text.set_text(text.get_text().replace(
"micro-average Precision-recall curve", "micro-avg"))
text.set_text(text.get_text().replace("macro-average Precision-recall",
"macro-avg"))
if not combine:
#plt.gcf().set_size_inches(3.65,3.65)
save_show(plt, library + "/" + prefix, sessionid,
"precision_recall_curve", show, save, False, True, True,
False)
else:
plt.subplot(2, 4, met_index + 1)
met_index += 1
if len(labels) < 3:
pltmetrics.plot_cumulative_gain(y_true, y_proba)
if not combine:
#plt.gcf().set_size_inches(3.65,3.65)
save_show(plt, library + "/" + prefix, sessionid,
"cumulative_gain", show, save, False, True, True, False)
else:
plt.subplot(2, 4, met_index + 1)
met_index += 1
if len(labels) < 3:
pltmetrics.plot_lift_curve(y_true, y_proba)
if not combine:
#plt.gcf().set_size_inches(3.65,3.65)
save_show(plt, library + "/" + prefix, sessionid, "lift_curve",
show, save, False, True, True, False)
else:
plt.subplot(2, 4, met_index + 1)
met_index += 1
if combine:
plt.suptitle(test_name)
plt.tight_layout(rect=[0, 0.03, 1, 0.95])
save_show(plt,
library,
sessionid,
figname,
show,
save,
True,
analysis=True)
def test_array_like(self):
plot_roc_curve([0, 'a'], [[0.8, 0.2], [0.2, 0.8]])
plot_roc_curve([0, 1], [[0.8, 0.2], [0.2, 0.8]])
plot_roc_curve(['b', 'a'], [[0.8, 0.2], [0.2, 0.8]])
def test_string_classes(self):
np.random.seed(0)
clf = LogisticRegression()
clf.fit(self.X, convert_labels_into_string(self.y))
probas = clf.predict_proba(self.X)
plot_roc_curve(convert_labels_into_string(self.y), probas)
print('Confusion matrix, without normalization')
print(cm)
plt.imshow(cm, interpolation='nearest', cmap=cmap)
plt.title(title)
plt.colorbar()
tick_marks = np.arange(len(classes))
plt.xticks(tick_marks, classes, rotation=45)
plt.yticks(tick_marks, classes)
fmt = '.2f' if normalize else 'd'
thresh = cm.max() / 2.
for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):
plt.text(j,
i,
format(cm[i, j], fmt),
horizontalalignment="center",
color="white" if cm[i, j] > thresh else "black")
plt.tight_layout()
plt.ylabel('True label')
plt.xlabel('Predicted label')
plt.figure()
plot_confusion_matrix(conf_matrix, classes, title='Matrice de confusion')
plt.show()
skplt.plot_roc_curve(trueTarget, predicted_proba)
plt.show()