Skip to content

Commit d2f9df2

Browse files
SarailidisSarailidis
authored
Add files via upload
1 parent 0ea8e94 commit d2f9df2

File tree

1 file changed

+118
-115
lines changed

1 file changed

+118
-115
lines changed

‎InteractiveDT/iDTGUIfun.py

Lines changed: 118 additions & 115 deletions
Original file line numberDiff line numberDiff line change
@@ -1244,121 +1244,124 @@ def calculate_accuracy(calcacc):
12441244
def plot_confusion_matrix(pltcm):
12451245
ConfusionMatrix_out.clear_output()
12461246
with ConfusionMatrix_out:
1247-
if Tree_state_widget.value == 'No expert tree interactions':
1248-
tree_orig=tree.DecisionTreeClassifier(criterion=inter_3.widget.kwargs['criterion'], splitter=inter_3.widget.kwargs['splitter'], max_depth=inter_3.widget.kwargs['max_depth'],
1249-
min_samples_split=inter_3.widget.kwargs['min_samples_split'], min_samples_leaf=inter_3.widget.kwargs['min_samples_leaf'],
1250-
min_weight_fraction_leaf=0.0, max_features=inter_3.widget.kwargs['max_features'], random_state=inter_3.widget.kwargs['random_state'],
1251-
max_leaf_nodes=inter_3.widget.kwargs['max_leaf_nodes'], min_impurity_decrease=inter_3.widget.kwargs['min_impurity_decrease'],
1252-
min_impurity_split=None, class_weight=None,ccp_alpha=0.0)
1253-
tree_orig.fit(self.Data_dict['x'],self.Data_dict['y'])
1254-
if self.Train_test_splitting.value == True:
1255-
CM_train=confusion_matrix(self.Data_dict['y'], tree_orig.predict(self.Data_dict['x']), labels=self.classes_dict['Classes Labels'])
1256-
CM_train_display=ConfusionMatrixDisplay(CM_train, self.classes_dict['Classes Labels'])
1257-
CM_test=confusion_matrix(self.Data_dict['w'], tree_orig.predict(self.Data_dict['z']), labels=self.classes_dict['Classes Labels'])
1258-
CM_test_display=ConfusionMatrixDisplay(CM_test, self.classes_dict['Classes Labels'])
1259-
CM_display_dict = {'Train': CM_train_display, 'Test': CM_test_display}
1260-
f, axes = plt.subplots(1, 2, figsize=(20, 5), sharey='row')
1261-
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1262-
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1263-
CM_display_dict.ax_.set_title(key)
1264-
CM_display_dict.im_.colorbar.remove()
1265-
CM_display_dict.ax_.set_xlabel('Predicted')
1266-
if i!=0:
1267-
CM_display_dict.ax_.set_ylabel('')
1268-
plt.subplots_adjust(wspace=0.40, hspace=0.1)
1269-
f.colorbar(CM_display_dict.im_, ax=axes)
1270-
plt.show()
1271-
elif self.Train_test_splitting.value == False:
1272-
CM=confusion_matrix(self.Data_dict['y'], tree_orig.predict(self.Data_dict['x']), labels=self.classes_dict['Classes Labels'])
1273-
CM_display=ConfusionMatrixDisplay(CM, self.classes_dict['Classes Labels'])
1274-
CM_display_dict = {'Confusion Matrix': CM_display}
1275-
f, axes = plt.subplots(1, 1, figsize=(20, 5))
1276-
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1277-
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1278-
CM_display_dict.ax_.set_title(key)
1279-
CM_display_dict.im_.colorbar.remove()
1280-
CM_display_dict.ax_.set_xlabel('Predicted')
1281-
f.colorbar(CM_display_dict.im_, ax=axes)
1282-
plt.show()
1283-
elif Tree_state_widget.value == 'Tree was last modified':
1284-
if self.Train_test_splitting.value == True:
1285-
#Make predictions on train set:
1286-
predicted_iDT_train=iDT.classify(base.Last_modified_tree, self.Data_dict['x'])
1287-
predicted_iDT_train.classify()
1288-
#Make predictions on test set:
1289-
predicted_iDT_test=iDT.classify(base.Last_modified_tree, self.Data_dict['z'])
1290-
predicted_iDT_test.classify()
1291-
#Calculate confusion Matrices:
1292-
CM_train=confusion_matrix(self.Data_dict['y'], predicted_iDT_train.predicted_classes, labels=self.classes_dict['Classes Labels'])
1293-
CM_train_display=ConfusionMatrixDisplay(CM_train, self.classes_dict['Classes Labels'])
1294-
CM_test=confusion_matrix(self.Data_dict['w'], predicted_iDT_test.predicted_classes, labels=self.classes_dict['Classes Labels'])
1295-
CM_test_display=ConfusionMatrixDisplay(CM_test, self.classes_dict['Classes Labels'])
1296-
CM_display_dict = {'Train': CM_train_display, 'Test': CM_test_display}
1297-
f, axes = plt.subplots(1, 2, figsize=(20, 5), sharey='row')
1298-
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1299-
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1300-
CM_display_dict.ax_.set_title(key)
1301-
CM_display_dict.im_.colorbar.remove()
1302-
CM_display_dict.ax_.set_xlabel('Predicted')
1303-
if i!=0:
1304-
CM_display_dict.ax_.set_ylabel('')
1305-
plt.subplots_adjust(wspace=0.40, hspace=0.1)
1306-
f.colorbar(CM_display_dict.im_, ax=axes)
1307-
plt.show()
1308-
elif self.Train_test_splitting.value == False:
1309-
predicted_iDT=iDT.classify(base.Last_modified_tree, self.Data_dict['x'])
1310-
predicted_iDT.classify()
1311-
#Calculate confusion Matrices:
1312-
CM=confusion_matrix(self.Data_dict['y'], predicted_iDT.predicted_classes, labels=self.classes_dict['Classes Labels'])
1313-
CM_display=ConfusionMatrixDisplay(CM, self.classes_dict['Classes Labels'])
1314-
CM_display_dict = {'Confusion Matrix': CM_display}
1315-
f, axes = plt.subplots(1, 1, figsize=(20, 5))
1316-
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1317-
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1318-
CM_display_dict.ax_.set_title(key)
1319-
CM_display_dict.im_.colorbar.remove()
1320-
CM_display_dict.ax_.set_xlabel('Predicted')
1321-
f.colorbar(CM_display_dict.im_, ax=axes)
1322-
plt.show()
1323-
elif Tree_state_widget.value == 'Tree was last pruned':
1324-
if self.Train_test_splitting.value == True:
1325-
#Make predictions on train set:
1326-
predicted_iDT_train=iDT.classify(base_prn.Last_Pruned_tree, self.Data_dict['x'])
1327-
predicted_iDT_train.classify()
1328-
#Make predictions on test set:
1329-
predicted_iDT_test=iDT.classify(base_prn.Last_Pruned_tree, self.Data_dict['z'])
1330-
predicted_iDT_test.classify()
1331-
#Calculate confusion Matrices:
1332-
CM_train=confusion_matrix(self.Data_dict['y'], predicted_iDT_train.predicted_classes, labels=self.classes_dict['Classes Labels'])
1333-
CM_train_display=ConfusionMatrixDisplay(CM_train, self.classes_dict['Classes Labels'])
1334-
CM_test=confusion_matrix(self.Data_dict['w'], predicted_iDT_test.predicted_classes, labels=self.classes_dict['Classes Labels'])
1335-
CM_test_display=ConfusionMatrixDisplay(CM_test, self.classes_dict['Classes Labels'])
1336-
CM_display_dict = {'Train': CM_train_display, 'Test': CM_test_display}
1337-
f, axes = plt.subplots(1, 2, figsize=(20, 5), sharey='row')
1338-
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1339-
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1340-
CM_display_dict.ax_.set_title(key)
1341-
CM_display_dict.im_.colorbar.remove()
1342-
CM_display_dict.ax_.set_xlabel('Predicted')
1343-
if i!=0:
1344-
CM_display_dict.ax_.set_ylabel('')
1345-
plt.subplots_adjust(wspace=0.40, hspace=0.1)
1346-
f.colorbar(CM_display_dict.im_, ax=axes)
1347-
plt.show()
1348-
elif self.Train_test_splitting.value == False:
1349-
predicted_iDT=iDT.classify(base_prn.Last_Pruned_tree, self.Data_dict['x'])
1350-
predicted_iDT.classify()
1351-
CM=confusion_matrix(self.Data_dict['y'], predicted_iDT.predicted_classes, labels=self.classes_dict['Classes Labels'])
1352-
CM_display=ConfusionMatrixDisplay(CM, self.classes_dict['Classes Labels'])
1353-
CM_display_dict = {'Confusion Matrix': CM_display}
1354-
f, axes = plt.subplots(1, 1, figsize=(20, 5))
1355-
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1356-
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1357-
CM_display_dict.ax_.set_title(key)
1358-
CM_display_dict.im_.colorbar.remove()
1359-
CM_display_dict.ax_.set_xlabel('Predicted')
1360-
f.colorbar(CM_display_dict.im_, ax=axes)
1361-
plt.show()
1247+
if len(self.classes_dict['Classes Labels']) == 0:
1248+
print("You havent assigned labels to classes. To plot a confusion matrix please first assign labels to classes using the appropriate tool in the Preprocessing Stage tab.")
1249+
else:
1250+
if Tree_state_widget.value == 'No expert tree interactions':
1251+
tree_orig=tree.DecisionTreeClassifier(criterion=inter_3.widget.kwargs['criterion'], splitter=inter_3.widget.kwargs['splitter'], max_depth=inter_3.widget.kwargs['max_depth'],
1252+
min_samples_split=inter_3.widget.kwargs['min_samples_split'], min_samples_leaf=inter_3.widget.kwargs['min_samples_leaf'],
1253+
min_weight_fraction_leaf=0.0, max_features=inter_3.widget.kwargs['max_features'], random_state=inter_3.widget.kwargs['random_state'],
1254+
max_leaf_nodes=inter_3.widget.kwargs['max_leaf_nodes'], min_impurity_decrease=inter_3.widget.kwargs['min_impurity_decrease'],
1255+
min_impurity_split=None, class_weight=None,ccp_alpha=0.0)
1256+
tree_orig.fit(self.Data_dict['x'],self.Data_dict['y'])
1257+
if self.Train_test_splitting.value == True:
1258+
CM_train=confusion_matrix(self.Data_dict['y'], tree_orig.predict(self.Data_dict['x']), labels=self.classes_dict['Classes Labels'])
1259+
CM_train_display=ConfusionMatrixDisplay(CM_train, self.classes_dict['Classes Labels'])
1260+
CM_test=confusion_matrix(self.Data_dict['w'], tree_orig.predict(self.Data_dict['z']), labels=self.classes_dict['Classes Labels'])
1261+
CM_test_display=ConfusionMatrixDisplay(CM_test, self.classes_dict['Classes Labels'])
1262+
CM_display_dict = {'Train': CM_train_display, 'Test': CM_test_display}
1263+
f, axes = plt.subplots(1, 2, figsize=(20, 5), sharey='row')
1264+
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1265+
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1266+
CM_display_dict.ax_.set_title(key)
1267+
CM_display_dict.im_.colorbar.remove()
1268+
CM_display_dict.ax_.set_xlabel('Predicted')
1269+
if i!=0:
1270+
CM_display_dict.ax_.set_ylabel('')
1271+
plt.subplots_adjust(wspace=0.40, hspace=0.1)
1272+
f.colorbar(CM_display_dict.im_, ax=axes)
1273+
plt.show()
1274+
elif self.Train_test_splitting.value == False:
1275+
CM=confusion_matrix(self.Data_dict['y'], tree_orig.predict(self.Data_dict['x']), labels=self.classes_dict['Classes Labels'])
1276+
CM_display=ConfusionMatrixDisplay(CM, self.classes_dict['Classes Labels'])
1277+
CM_display_dict = {'Confusion Matrix': CM_display}
1278+
f, axes = plt.subplots(1, 1, figsize=(20, 5))
1279+
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1280+
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1281+
CM_display_dict.ax_.set_title(key)
1282+
CM_display_dict.im_.colorbar.remove()
1283+
CM_display_dict.ax_.set_xlabel('Predicted')
1284+
f.colorbar(CM_display_dict.im_, ax=axes)
1285+
plt.show()
1286+
elif Tree_state_widget.value == 'Tree was last modified':
1287+
if self.Train_test_splitting.value == True:
1288+
#Make predictions on train set:
1289+
predicted_iDT_train=iDT.classify(base.Last_modified_tree, self.Data_dict['x'])
1290+
predicted_iDT_train.classify()
1291+
#Make predictions on test set:
1292+
predicted_iDT_test=iDT.classify(base.Last_modified_tree, self.Data_dict['z'])
1293+
predicted_iDT_test.classify()
1294+
#Calculate confusion Matrices:
1295+
CM_train=confusion_matrix(self.Data_dict['y'], predicted_iDT_train.predicted_classes, labels=self.classes_dict['Classes Labels'])
1296+
CM_train_display=ConfusionMatrixDisplay(CM_train, self.classes_dict['Classes Labels'])
1297+
CM_test=confusion_matrix(self.Data_dict['w'], predicted_iDT_test.predicted_classes, labels=self.classes_dict['Classes Labels'])
1298+
CM_test_display=ConfusionMatrixDisplay(CM_test, self.classes_dict['Classes Labels'])
1299+
CM_display_dict = {'Train': CM_train_display, 'Test': CM_test_display}
1300+
f, axes = plt.subplots(1, 2, figsize=(20, 5), sharey='row')
1301+
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1302+
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1303+
CM_display_dict.ax_.set_title(key)
1304+
CM_display_dict.im_.colorbar.remove()
1305+
CM_display_dict.ax_.set_xlabel('Predicted')
1306+
if i!=0:
1307+
CM_display_dict.ax_.set_ylabel('')
1308+
plt.subplots_adjust(wspace=0.40, hspace=0.1)
1309+
f.colorbar(CM_display_dict.im_, ax=axes)
1310+
plt.show()
1311+
elif self.Train_test_splitting.value == False:
1312+
predicted_iDT=iDT.classify(base.Last_modified_tree, self.Data_dict['x'])
1313+
predicted_iDT.classify()
1314+
#Calculate confusion Matrices:
1315+
CM=confusion_matrix(self.Data_dict['y'], predicted_iDT.predicted_classes, labels=self.classes_dict['Classes Labels'])
1316+
CM_display=ConfusionMatrixDisplay(CM, self.classes_dict['Classes Labels'])
1317+
CM_display_dict = {'Confusion Matrix': CM_display}
1318+
f, axes = plt.subplots(1, 1, figsize=(20, 5))
1319+
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1320+
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1321+
CM_display_dict.ax_.set_title(key)
1322+
CM_display_dict.im_.colorbar.remove()
1323+
CM_display_dict.ax_.set_xlabel('Predicted')
1324+
f.colorbar(CM_display_dict.im_, ax=axes)
1325+
plt.show()
1326+
elif Tree_state_widget.value == 'Tree was last pruned':
1327+
if self.Train_test_splitting.value == True:
1328+
#Make predictions on train set:
1329+
predicted_iDT_train=iDT.classify(base_prn.Last_Pruned_tree, self.Data_dict['x'])
1330+
predicted_iDT_train.classify()
1331+
#Make predictions on test set:
1332+
predicted_iDT_test=iDT.classify(base_prn.Last_Pruned_tree, self.Data_dict['z'])
1333+
predicted_iDT_test.classify()
1334+
#Calculate confusion Matrices:
1335+
CM_train=confusion_matrix(self.Data_dict['y'], predicted_iDT_train.predicted_classes, labels=self.classes_dict['Classes Labels'])
1336+
CM_train_display=ConfusionMatrixDisplay(CM_train, self.classes_dict['Classes Labels'])
1337+
CM_test=confusion_matrix(self.Data_dict['w'], predicted_iDT_test.predicted_classes, labels=self.classes_dict['Classes Labels'])
1338+
CM_test_display=ConfusionMatrixDisplay(CM_test, self.classes_dict['Classes Labels'])
1339+
CM_display_dict = {'Train': CM_train_display, 'Test': CM_test_display}
1340+
f, axes = plt.subplots(1, 2, figsize=(20, 5), sharey='row')
1341+
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1342+
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1343+
CM_display_dict.ax_.set_title(key)
1344+
CM_display_dict.im_.colorbar.remove()
1345+
CM_display_dict.ax_.set_xlabel('Predicted')
1346+
if i!=0:
1347+
CM_display_dict.ax_.set_ylabel('')
1348+
plt.subplots_adjust(wspace=0.40, hspace=0.1)
1349+
f.colorbar(CM_display_dict.im_, ax=axes)
1350+
plt.show()
1351+
elif self.Train_test_splitting.value == False:
1352+
predicted_iDT=iDT.classify(base_prn.Last_Pruned_tree, self.Data_dict['x'])
1353+
predicted_iDT.classify()
1354+
CM=confusion_matrix(self.Data_dict['y'], predicted_iDT.predicted_classes, labels=self.classes_dict['Classes Labels'])
1355+
CM_display=ConfusionMatrixDisplay(CM, self.classes_dict['Classes Labels'])
1356+
CM_display_dict = {'Confusion Matrix': CM_display}
1357+
f, axes = plt.subplots(1, 1, figsize=(20, 5))
1358+
for i, (key, CM_display_dict) in enumerate(CM_display_dict.items()):
1359+
CM_display_dict.plot(ax=axes[i], xticks_rotation=45)
1360+
CM_display_dict.ax_.set_title(key)
1361+
CM_display_dict.im_.colorbar.remove()
1362+
CM_display_dict.ax_.set_xlabel('Predicted')
1363+
f.colorbar(CM_display_dict.im_, ax=axes)
1364+
plt.show()
13621365

13631366
#Assign the above function to calculate accuracy button click
13641367
ConfusionMatrix_button.on_click(plot_confusion_matrix)

0 commit comments

Comments
 (0)