Package org.apache.ignite.examples.ml.tutorial
Full ML Tutorial.
-
Class Summary Class Description Step_1_Read_and_Learn Usage ofDecisionTreeClassificationTrainerto predict death in the disaster.Step_10_Bagging MinMaxScalerTrainerandNormalizationTrainerare used in this example due to different values distribution in columns and rows.Step_11_Boosting MinMaxScalerTrainerandNormalizationTrainerare used in this example due to different values distribution in columns and rows.Step_12_Model_Update MinMaxScalerTrainerandNormalizationTrainerare used in this example due to different values distribution in columns and rows.Step_2_Imputing Usage ofImputerTrainerto fill missed data (Double.NaN) values in the chosen columns.Step_3_Categorial Let's add two categorial features "sex", "embarked" to predict more precisely than inStep_1_Read_and_Learn.Step_3_Categorial_with_One_Hot_Encoder Let's add two categorial features "sex", "embarked" to predict more precisely than inStep_1_Read_and_Learn..Step_4_Add_age_fare Add yet two numerical features "age", "fare" to improve our model overStep_3_Categorial.Step_5_Scaling MinMaxScalerTrainerandNormalizationTrainerare used in this example due to different values distribution in columns and rows.Step_5_Scaling_with_Pipeline MinMaxScalerTrainerandNormalizationTrainerare used in this example due to different values distribution in columns and rows.Step_6_KNN Change classification algorithm that was used inStep_5_Scalingfrom decision tree to kNN (KNNClassificationTrainer) because sometimes this can be beneficial.Step_7_Split_train_test The highest accuracy in the previous example (Step_6_KNN) is the result of overfitting.Step_8_CV To choose the best hyper-parameters the cross-validation will be used in this example.Step_8_CV_with_Param_Grid To choose the best hyper-parameters the cross-validation withParamGridwill be used in this example.Step_8_CV_with_Param_Grid_and_pipeline To choose the best hyper-parameters the cross-validation withParamGridwill be used in this example.Step_9_Scaling_With_Stacking MinMaxScalerTrainerandNormalizationTrainerare used in this example due to different values distribution in columns and rows.TitanicUtils The utility class.