def multi_classification(): digits = load_digits() x = digits.data y = digits.target sample_parameter = { 'n_jobs': -1, 'min_samples_leaf': 2.0, 'n_estimators': 500, 'max_features': 0.55, 'criterion': 'mse', 'min_samples_split': 4.0, 'model': 'RFCLF', 'max_depth': 4.0 } x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.2,random_state=42) clf_layer = mlc.layer.layer.ClassificationLayer() print "single prediction" y_train_predict,y_test_predict = clf_layer.predict(x_train,y_train,x_test,sample_parameter) #print y_test_predict y_train_predict_proba,y_test_predict_proba = clf_layer.predict_proba(x_train,y_train,x_test,sample_parameter) #print y_test_predict_proba print evaluate_function(y_test,np.argmax(y_test_predict_proba,axis=1),'accuracy') print "multi ensamble prediction" multi_bagging_clf = mlc.layer.layer.ClassificationMultiBaggingLayer() y_train_predict_proba,y_test_predict_proba = multi_bagging_clf.predict_proba(x_train,y_train,x_test,sample_parameter,times=5) print evaluate_function(y_test,np.argmax(y_test_predict_proba,axis=1),'accuracy')
def bagging_regression(): digits = load_diabetes() x = digits.data y = digits.target sample_parameter = { 'n_jobs': -1, 'min_samples_leaf': 2.0, 'n_estimators': 500, 'max_features': 0.55, 'criterion': 'mse', 'min_samples_split': 4.0, 'model': 'RFREG', 'max_depth': 4.0 } x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.2,random_state=42) clf_layer = mlc.layer.layer.RegressionLayer() print "single prediction" #y_train_predict,y_test_predict = clf_layer.predict(x_train,y_train,x_test,sample_parameter) #print y_test_predict y_train_predict_proba,y_test_predict_proba = clf_layer.predict(x_train,y_train,x_test,sample_parameter) #print y_test_predict_proba print evaluate_function(y_test,y_test_predict_proba,'mean_squared_error') print "multi ensamble prediction" multi_bagging_clf = mlc.layer.layer.RegressionBaggingLayer() y_train_predict_proba,y_test_predict_proba = multi_bagging_clf.predict(x_train,y_train,x_test,sample_parameter,times=5) print evaluate_function(y_test,y_test_predict_proba,'mean_squared_error')
def binary_classification(): digits = load_digits(n_class=2) x = digits.data y = digits.target sample_parameter = { "colsample_bytree": 0.9, "min_child_weight": 10, "num_round": 300, "subsample": 0.7, "eta": 0.2, "max_depth": 4, "gamma": 0.6000000000000001, "model": "XGBREGLOGISTIC", "objective": "binary:logistic" } x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.2,random_state=42) clf_layer = mlc.layer.layer.ClassificationBinaryLayer() print "single prediction" #y_train_predict,y_test_predict = clf_layer.predict(x_train,y_train,x_test,sample_parameter) #print y_test_predict y_train_predict_proba,y_test_predict_proba = clf_layer.predict_proba(x_train,y_train,x_test,sample_parameter) #print y_test_predict_proba print evaluate_function(y_test,y_test_predict_proba,'auc') print "multi ensamble prediction" multi_bagging_clf = mlc.layer.layer.ClassificationBinaryBaggingLayer() y_train_predict_proba,y_test_predict_proba = multi_bagging_clf.predict_proba(x_train,y_train,x_test,sample_parameter,times=5) print evaluate_function(y_test,y_test_predict_proba,'auc')
def multi_classification(): digits = load_digits() x = digits.data y = digits.target sample_parameter = { 'n_jobs': -1, 'min_samples_leaf': 2.0, 'n_estimators': 500, 'max_features': 0.55, 'criterion': 'mse', 'min_samples_split': 4.0, 'model': 'RFCLF', 'max_depth': 4.0 } x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=42) clf_layer = mlc.layer.layer.ClassificationLayer() print "single prediction" y_train_predict, y_test_predict = clf_layer.predict( x_train, y_train, x_test, sample_parameter) #print y_test_predict y_train_predict_proba, y_test_predict_proba = clf_layer.predict_proba( x_train, y_train, x_test, sample_parameter) #print y_test_predict_proba print evaluate_function(y_test, np.argmax(y_test_predict_proba, axis=1), 'accuracy') print "multi ensamble prediction" multi_bagging_clf = mlc.layer.layer.ClassificationMultiBaggingLayer() y_train_predict_proba, y_test_predict_proba = multi_bagging_clf.predict_proba( x_train, y_train, x_test, sample_parameter, times=5) print evaluate_function(y_test, np.argmax(y_test_predict_proba, axis=1), 'accuracy')
def binary_classification(): digits = load_digits(n_class=2) x = digits.data y = digits.target sample_parameter = { "colsample_bytree": 0.9, "min_child_weight": 10, "num_round": 300, "subsample": 0.7, "eta": 0.2, "max_depth": 4, "gamma": 0.6000000000000001, "model": "XGBREGLOGISTIC", "objective": "binary:logistic" } x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=42) clf_layer = mlc.layer.layer.ClassificationBinaryLayer() print "single prediction" #y_train_predict,y_test_predict = clf_layer.predict(x_train,y_train,x_test,sample_parameter) #print y_test_predict y_train_predict_proba, y_test_predict_proba = clf_layer.predict_proba( x_train, y_train, x_test, sample_parameter) #print y_test_predict_proba print evaluate_function(y_test, y_test_predict_proba, 'auc') print "multi ensamble prediction" multi_bagging_clf = mlc.layer.layer.ClassificationBinaryBaggingLayer() y_train_predict_proba, y_test_predict_proba = multi_bagging_clf.predict_proba( x_train, y_train, x_test, sample_parameter, times=5) print evaluate_function(y_test, y_test_predict_proba, 'auc')
def bagging_regression(): digits = load_diabetes() x = digits.data y = digits.target sample_parameter = { 'n_jobs': -1, 'min_samples_leaf': 2.0, 'n_estimators': 500, 'max_features': 0.55, 'criterion': 'mse', 'min_samples_split': 4.0, 'model': 'RFREG', 'max_depth': 4.0 } x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=42) clf_layer = mlc.layer.layer.RegressionLayer() print "single prediction" #y_train_predict,y_test_predict = clf_layer.predict(x_train,y_train,x_test,sample_parameter) #print y_test_predict y_train_predict_proba, y_test_predict_proba = clf_layer.predict( x_train, y_train, x_test, sample_parameter) #print y_test_predict_proba print evaluate_function(y_test, y_test_predict_proba, 'mean_squared_error') print "multi ensamble prediction" multi_bagging_clf = mlc.layer.layer.RegressionBaggingLayer() y_train_predict_proba, y_test_predict_proba = multi_bagging_clf.predict( x_train, y_train, x_test, sample_parameter, times=5) print evaluate_function(y_test, y_test_predict_proba, 'mean_squared_error')