append (( "NB", GaussianNB ())) # uses a threshold that is the midpoint between different classifications all_models. append (( "CART", DecisionTreeClassifier ())) # uses the Gaussian distribution to classify new data all_models. append (( "KNN", KNeighborsClassifier ())) # tree-like diagram where each leaf node is the outcome that is used to classify new data all_models.
append (( "LDA", LinearDiscriminantAnalysis ())) # classification algorithm that memorizes observations to classify new data # - new data is classified by observing the "nearest neighbours" all_models. append (( "LogReg", LogisticRegression ( solver = "liblinear", multi_class = "ovr" ))) # maximizes seperation using our chosen features # - maximizing distance between means and minimizing variance # - reduces features down # - minimizes the scatter all_models. # Algorithms all_models = # liblinear has both L1 and L2 regularization # - Ridge Regression and Lasso Regression -> avoid overfitting and feature selection # ovr means it's a binary problem for each label (one vs rest) all_models.
Minutes Goals Assists Clean_Sheets Own_Goals \
0 kommentar(er)
