Machine Learning

For a complex question, single classifier or predictor may give way to a multiple of them, an aggregated answer is always better, think about your life experience. And today, popular ensemble methods including bagging, boosting and stacking. Random Forest is also prevailing. Voting Classifier We start with the simple intuition, building a voting system over different classifiers, they could be Logistic Regression, SVM Classifier and Decision Tree… Then we count each classifier’s prediction and to get a overall prediction, it can be the majority vote, and this mechanism is call hard voting, or we can add a weight to each classifier, and obviously this is soft voting. »

Classification Metrics Classification is about predicting class labels given input data. Take binary classification for example, there’re 2 possible classes, usually positive and negative, so we can naively use accuracy to measure the performance. $$Accuracy = {n\_correct \over n\_total}$$ In this competition, you’ll distinguish dogs from cats However, in reality, the classes are not always equal. The quantities of classes may be biased, and the cost of making wrong prediction may be different (e. »