Heatmaps.py

#6 Heatmaps -->
lr_cm = metrics.confusion_matrix(y_test, lr_predict) # cm = confusion matrix, so sklearn has a built in for that.
svc_cm = metrics.confusion_matrix(y_test, svc_predict)
rf_cm = metrics.confusion_matrix(y_test, rf_predict)
knn_cm = metrics.confusion_matrix(y_test, knn_predict)

# Logistic regression
plt.figure(figsize=(9,9))
sns.heatmap(lr_cm, annot=True, fmt='.3f', linewidths=.5, square=True, cmap='Blues');
plt.ylabel('Actual label');
plt.xlabel('Predicted label');
all_sample_title = 'Logistic Regression: {0}'.format(lr_score)
plt.title(all_sample_title, size=15);

# Support Vector Machines -->
plt.figure(figsize=(9,9))
sns.heatmap(svc_cm, annot=True, fmt='.3f', linewidths=.5, square=True, cmap='Blues');
plt.ylabel('Actual label');
plt.xlabel('Predicted label');
all_sample_title = 'Support Vector Machines: {0}'.format(svc_score)
plt.title(all_sample_title, size=15);

# Random Forest Classifier -->
plt.figure(figsize=(9,9))
sns.heatmap(rf_cm, annot=True, fmt='.3f', linewidths=.5, square=True, cmap='Blues');
plt.ylabel('Actual label');
plt.xlabel('Predicted label');
all_sample_title = 'Random Forest: {0}'.format(svc_score)
plt.title(all_sample_title, size=15);

# K-Nearest Neighbors -->
plt.figure(figsize=(9,9))
sns.heatmap(knn_cm, annot=True, fmt='.3f', linewidths=.5, square=True, cmap='Blues');
plt.ylabel('Actual label');
plt.xlabel('Predicted label');
all_sample_title = 'K-Nearest Neighbors: {0}'.format(svc_score)
plt.title(all_sample_title, size=15);

# Printout plots -->
plt.show()